Codebase intelligence for AI and humans: code health scores, auto-generated docs, git analytics, dead code detection, and architectural decisions via MCP.
Python aicode-complexitycode-healthcode-intelligencecode-qualitydead-codedeveloper-toolsdocumentationgit-analyticsmcpopen-sourcepythonrefactoringstatic-analysistechnical-debt
Scope the Stats page to what no other page shows (#1091)
Raghav Chamadiya65aa1c43f5ee
diff --git a/packages/api-client/src/__fixtures__/hosted/stats-highlights.json b/packages/api-client/src/__fixtures__/hosted/stats-highlights.jsonindex 3dbe15593..b86c54958 100644--- a/packages/api-client/src/__fixtures__/hosted/stats-highlights.json+++ b/packages/api-client/src/__fixtures__/hosted/stats-highlights.json@@ -1,1157 +1,379 @@{"repo": {- "id": "d8ea204a-6325-438c-a805-fbdb029ffd11",- "name": "flask",- "default_branch": "main",- "head_commit": "36e4a824f340fdee7ed50937ba8e7f6bc7d17f81"+ "id": "repo-1",+ "name": "demo"},"scale": {- "file_count": 108,- "symbol_count": 1060,- "entry_point_count": 5,- "module_count": 24,- "total_nloc": 14043,- "language_count": 8,+ "file_count": 3146,+ "symbol_count": 24430,+ "module_count": 11,+ "total_nloc": 419590,+ "language_count": 12,"languages": [{"language": "python",- "file_count": 83+ "file_count": 1803},{- "language": "yaml",- "file_count": 8+ "language": "typescript",+ "file_count": 1001},{"language": "markdown",- "file_count": 6- },- {- "language": "toml",- "file_count": 5+ "file_count": 144},{"language": "json",- "file_count": 2- },- {- "language": "sql",- "file_count": 2- },- {- "language": "shell",- "file_count": 1- },- {- "language": "makefile",- "file_count": 1+ "file_count": 85}],"size_class": {- "name": "Village",- "blurb": "A tidy village — a few neighborhoods, easy to walk.",- "nloc": 14043+ "name": "Metropolis",+ "blurb": "A sprawling metropolis with serious infrastructure.",+ "nloc": 419590}},- "activity": {- "total_commits": 2000,- "agent_commits": 0,- "agent_pct": 0,- "fix_commits": 204,- "fix_pct": 10.2,- "contributor_count": 568,- "first_commit_at": "2015-06-06T04:25:41+00:00",- "first_commit_author": "Armin Ronacher",- "last_commit_at": "2026-05-31T14:42:46+00:00",- "age_days": 4012,- "busiest_month": {- "month": "2017-05",- "total": 100,- "agent": 0- },- "monthly": [- {- "month": "2015-06",- "total": 22,- "agent": 0- },- {- "month": "2015-07",- "total": 12,- "agent": 0- },- {- "month": "2015-08",- "total": 3,- "agent": 0- },- {- "month": "2015-09",- "total": 2,- "agent": 0- },- {- "month": "2015-10",- "total": 10,- "agent": 0- },- {- "month": "2015-11",- "total": 8,- "agent": 0- },- {- "month": "2015-12",- "total": 10,- "agent": 0- },- {- "month": "2016-01",- "total": 6,- "agent": 0- },- {- "month": "2016-02",- "total": 6,- "agent": 0- },- {- "month": "2016-03",- "total": 7,- "agent": 0- },- {- "month": "2016-04",- "total": 18,- "agent": 0- },- {- "month": "2016-05",- "total": 33,- "agent": 0- },- {- "month": "2016-06",- "total": 71,- "agent": 0- },- {- "month": "2016-07",- "total": 9,- "agent": 0- },- {- "month": "2016-08",- "total": 33,- "agent": 0- },- {- "month": "2016-09",- "total": 22,- "agent": 0- },- {- "month": "2016-10",- "total": 10,- "agent": 0- },- {- "month": "2016-11",- "total": 13,- "agent": 0- },- {- "month": "2016-12",- "total": 21,- "agent": 0- },- {- "month": "2017-01",- "total": 10,- "agent": 0- },- {- "month": "2017-02",- "total": 10,- "agent": 0- },- {- "month": "2017-03",- "total": 27,- "agent": 0- },- {- "month": "2017-04",- "total": 29,- "agent": 0- },- {- "month": "2017-05",- "total": 100,- "agent": 0- },- {- "month": "2017-06",- "total": 54,- "agent": 0- },- {- "month": "2017-07",- "total": 11,- "agent": 0- },- {- "month": "2017-08",- "total": 9,- "agent": 0- },- {- "month": "2017-09",- "total": 4,- "agent": 0- },- {- "month": "2017-10",- "total": 9,- "agent": 0- },- {- "month": "2017-11",- "total": 8,- "agent": 0- },- {- "month": "2017-12",- "total": 16,- "agent": 0- },- {- "month": "2018-01",- "total": 25,- "agent": 0- },- {- "month": "2018-02",- "total": 14,- "agent": 0- },- {- "month": "2018-03",- "total": 4,- "agent": 0- },- {- "month": "2018-04",- "total": 77,- "agent": 0- },- {- "month": "2018-05",- "total": 27,- "agent": 0- },- {- "month": "2018-06",- "total": 11,- "agent": 0- },- {- "month": "2018-07",- "total": 4,- "agent": 0- },- {- "month": "2018-08",- "total": 1,- "agent": 0- },- {- "month": "2018-09",- "total": 12,- "agent": 0- },- {- "month": "2018-10",- "total": 6,- "agent": 0- },- {- "month": "2018-11",- "total": 7,- "agent": 0- },- {- "month": "2018-12",- "total": 2,- "agent": 0- },- {- "month": "2019-01",- "total": 36,- "agent": 0- },- {- "month": "2019-02",- "total": 2,- "agent": 0- },- {- "month": "2019-03",- "total": 1,- "agent": 0- },- {- "month": "2019-04",- "total": 2,- "agent": 0- },- {- "month": "2019-05",- "total": 73,- "agent": 0- },- {- "month": "2019-06",- "total": 26,- "agent": 0- },- {- "month": "2019-07",- "total": 19,- "agent": 0- },- {- "month": "2019-08",- "total": 8,- "agent": 0- },- {- "month": "2019-09",- "total": 5,- "agent": 0- },- {- "month": "2019-10",- "total": 10,- "agent": 0- },- {- "month": "2019-11",- "total": 11,- "agent": 0- },- {- "month": "2020-02",- "total": 22,- "agent": 0- },- {- "month": "2020-03",- "total": 8,- "agent": 0- },- {- "month": "2020-04",- "total": 22,- "agent": 0- },- {- "month": "2020-05",- "total": 8,- "agent": 0- },- {- "month": "2020-06",- "total": 19,- "agent": 0- },- {- "month": "2020-07",- "total": 33,- "agent": 0- },- {- "month": "2020-08",- "total": 12,- "agent": 0- },- {- "month": "2020-09",- "total": 6,- "agent": 0- },- {- "month": "2020-10",- "total": 10,- "agent": 0- },- {- "month": "2020-11",- "total": 21,- "agent": 0- },- {- "month": "2020-12",- "total": 2,- "agent": 0- },- {- "month": "2021-01",- "total": 16,- "agent": 0- },- {- "month": "2021-02",- "total": 26,- "agent": 0- },- {- "month": "2021-03",- "total": 14,- "agent": 0- },- {- "month": "2021-04",- "total": 34,- "agent": 0- },- {- "month": "2021-05",- "total": 67,- "agent": 0- },- {- "month": "2021-06",- "total": 15,- "agent": 0- },- {- "month": "2021-07",- "total": 12,- "agent": 0- },- {- "month": "2021-08",- "total": 21,- "agent": 0- },- {- "month": "2021-09",- "total": 10,- "agent": 0- },- {- "month": "2021-10",- "total": 14,- "agent": 0- },- {- "month": "2021-11",- "total": 24,- "agent": 0- },- {- "month": "2021-12",- "total": 14,- "agent": 0- },- {- "month": "2022-01",- "total": 7,- "agent": 0- },- {- "month": "2022-02",- "total": 14,- "agent": 0- },- {- "month": "2022-03",- "total": 23,- "agent": 0- },- {- "month": "2022-04",- "total": 16,- "agent": 0- },- {- "month": "2022-05",- "total": 24,- "agent": 0- },- {- "month": "2022-06",- "total": 32,- "agent": 0- },- {- "month": "2022-07",- "total": 33,- "agent": 0- },- {- "month": "2022-08",- "total": 32,- "agent": 0- },- {- "month": "2022-09",- "total": 6,- "agent": 0- },- {- "month": "2022-10",- "total": 4,- "agent": 0- },- {- "month": "2022-11",- "total": 5,- "agent": 0- },- {- "month": "2022-12",- "total": 14,- "agent": 0- },- {- "month": "2023-01",- "total": 13,- "agent": 0- },- {- "month": "2023-02",- "total": 22,- "agent": 0- },- {- "month": "2023-03",- "total": 8,- "agent": 0- },- {- "month": "2023-04",- "total": 34,- "agent": 0- },- {- "month": "2023-05",- "total": 24,- "agent": 0- },- {- "month": "2023-06",- "total": 21,- "agent": 0- },- {- "month": "2023-07",- "total": 6,- "agent": 0- },- {- "month": "2023-08",- "total": 24,- "agent": 0- },- {- "month": "2023-09",- "total": 7,- "agent": 0- },- {- "month": "2023-10",- "total": 5,- "agent": 0- },- {- "month": "2023-11",- "total": 14,- "agent": 0- },- {- "month": "2023-12",- "total": 6,- "agent": 0- },- {- "month": "2024-01",- "total": 11,- "agent": 0- },- {- "month": "2024-02",- "total": 8,- "agent": 0- },- {- "month": "2024-04",- "total": 14,- "agent": 0- },- {- "month": "2024-05",- "total": 11,- "agent": 0- },- {- "month": "2024-06",- "total": 4,- "agent": 0- },- {- "month": "2024-07",- "total": 5,- "agent": 0- },- {- "month": "2024-08",- "total": 10,- "agent": 0- },- {- "month": "2024-09",- "total": 4,- "agent": 0- },- {- "month": "2024-10",- "total": 13,- "agent": 0- },- {- "month": "2024-11",- "total": 22,- "agent": 0- },- {- "month": "2025-01",- "total": 4,- "agent": 0- },- {- "month": "2025-03",- "total": 11,- "agent": 0- },- {- "month": "2025-05",- "total": 8,- "agent": 0- },- {- "month": "2025-06",- "total": 7,- "agent": 0- },- {- "month": "2025-08",- "total": 13,- "agent": 0- },- {- "month": "2025-09",- "total": 3,- "agent": 0- },- {- "month": "2025-10",- "total": 1,- "agent": 0- },- {- "month": "2025-11",- "total": 4,- "agent": 0- },- {- "month": "2026-01",- "total": 12,- "agent": 0- },- {- "month": "2026-02",- "total": 11,- "agent": 0- },- {- "month": "2026-03",- "total": 4,- "agent": 0- },- {- "month": "2026-04",- "total": 4,- "agent": 0- },- {- "month": "2026-05",- "total": 6,- "agent": 0- }- ],- "agent_names": [],+ "origin": {+ "first_commit_at": "2026-03-23T16:14:58+00:00",+ "first_commit_author": "RaghavChamadiya",+ "first_commit_subject": "Initial commit: CodeWiki project",+ "last_commit_at": "2026-07-25T18:00:21+00:00",+ "age_days": 124,+ "total_commits": 854,+ "contributor_count": 51+ },+ "churn": {+ "lines_added": 740846,+ "lines_deleted": 160171,+ "net": 580675,+ "deleted_per_hundred": 21.6+ },+ "rhythm": {"punch_card": {"matrix": [[- 1,- 1,- 1,- 1,- 1,- 1,- 1,- 6,+ 2,+ 0,+ 0,+ 0,+ 0,+ 0,+ 0,+ 0,+ 0,+ 1,+ 9,+ 5,6,+ 5,+ 8,12,- 12,- 12,- 12,- 12,- 12,- 12,- 12,- 12,- 12,- 6,- 6,- 6,- 1,- 1+ 10,+ 4,+ 4,+ 11,+ 5,+ 0,+ 3,+ 4],[+ 0,+ 0,+ 0,+ 0,+ 0,+ 0,+ 0,+ 0,+ 0,+ 0,+ 4,1,- 1,- 1,- 1,- 1,- 1,- 1,- 6,- 6,- 12,- 12,- 12,- 12,- 12,- 12,- 12,- 12,- 12,- 12,- 6,+ 3,+ 8,+ 13,+ 16,+ 4,+ 5,+ 5,6,+ 5,+ 3,6,- 1,- 1+ 2],[1,+ 0,+ 0,+ 0,+ 0,+ 0,+ 0,+ 0,1,1,- 1,- 1,- 1,- 1,- 6,- 6,- 12,- 12,- 12,- 12,- 12,- 12,- 12,- 12,- 12,- 12,- 6,- 6,- 6,- 1,- 1+ 4,+ 8,+ 7,+ 17,+ 12,+ 13,+ 11,+ 11,+ 7,+ 2,+ 16,+ 8,+ 4,+ 4],[- 1,- 1,- 1,- 1,- 1,- 1,- 1,- 6,- 6,- 12,- 12,- 12,- 12,- 12,- 12,- 12,- 12,- 12,+ 3,+ 0,+ 0,+ 0,+ 0,+ 0,+ 0,+ 0,+ 0,+ 0,+ 3,+ 9,+ 4,+ 0,+ 9,+ 16,+ 19,+ 15,+ 4,+ 9,12,6,- 6,- 6,1,- 1+ 7],[1,- 1,- 1,- 1,- 1,- 1,- 1,+ 2,+ 0,+ 0,+ 0,+ 0,+ 0,+ 0,+ 0,+ 3,6,+ 16,+ 10,+ 11,+ 25,+ 13,+ 13,+ 13,+ 7,6,- 12,- 12,- 12,- 12,- 12,- 12,- 12,- 12,- 12,- 12,- 6,- 6,- 6,- 1,- 1+ 8,+ 10,+ 4,+ 7],[- 1,- 1,- 1,- 1,- 1,- 1,- 1,- 1,- 1,- 1,- 1,- 4,4,+ 3,+ 0,+ 0,+ 0,+ 0,+ 0,+ 2,+ 3,+ 1,+ 9,+ 15,+ 7,+ 10,+ 18,+ 18,+ 16,+ 18,+ 15,+ 17,+ 7,4,4,- 4,- 4,- 1,- 1,- 1,- 1,- 1,- 1,- 1+ 4],[1,- 1,- 1,- 1,- 1,- 1,- 1,- 1,- 1,- 1,- 1,- 4,- 4,- 4,+ 0,+ 0,+ 0,+ 0,+ 0,+ 0,+ 0,+ 0,+ 1,+ 1,+ 9,+ 9,+ 8,+ 5,+ 21,+ 22,+ 10,+ 13,+ 2,4,- 4,- 4,- 1,+ 0,1,- 1,- 1,- 1,- 1,- 1+ 2]],"peak": {- "weekday": 2,+ "weekday": 4,"hour": 14,- "count": 12+ "count": 25},- "busiest_weekday": 2,- "peak_hour": 14,- "total": 879+ "busiest_weekday": 5,+ "peak_hour": 15,+ "total": 853,+ "timezone_mode": "author_local"},"velocity": {- "recent_90d": 412,- "prior_90d": 318,- "pct_change": 29.6+ "recent_90d": 742,+ "prior_90d": 111,+ "pct_change": 568.5},- "change_risk_mix": {- "low": 1240,- "moderate": 560,- "high": 200- }+ "busiest_month": {+ "month": "2026-07",+ "total": 351+ },+ "busiest_day": {+ "date": "2026-07-24",+ "commits": 37+ },+ "longest_streak": {+ "days": 25,+ "start": "2026-05-23",+ "end": "2026-06-16"+ },+ "active_days": 95,+ "code_half_life_days": 28},"people": {- "owner_count": 12,- "top_owners": [- {- "name": "David Lord",- "file_count": 58,- "pct": 0.6904761904761905- },- {- "name": "Markus Unterwaditzer",- "file_count": 5,- "pct": 0.05952380952380952- },- {- "name": "pgjones",- "file_count": 4,- "pct": 0.047619047619047616- },+ "owner_count": 28,+ "single_owner_files": 2820,+ "silo_count": 9,+ "truck_factor": 1,+ "contributor_count": 51,+ "chronotypes": [+ {+ "name": "Raghav Chamadiya",+ "commits": 626,+ "peak_hour": 15,+ "label": "daylight",+ "night_pct": 9.6,+ "early_pct": 0.8+ },+ {+ "name": "Swati Ahuja",+ "commits": 90,+ "peak_hour": 15,+ "label": "daylight",+ "night_pct": 10.0,+ "early_pct": 0.0+ },+ {+ "name": "joptimus",+ "commits": 25,+ "peak_hour": 13,+ "label": "daylight",+ "night_pct": 0.0,+ "early_pct": 0.0+ }+ ],+ "arrivals": [{- "name": "David Baumgold",- "file_count": 4,- "pct": 0.047619047619047616+ "name": "Raghav Chamadiya",+ "first_commit_at": "2026-03-23T09:14:58+00:00"},{- "name": "Armin Ronacher",- "file_count": 4,- "pct": 0.047619047619047616+ "name": "Swati Ahuja",+ "first_commit_at": "2026-03-29T07:34:55+00:00"},{- "name": "Jannis Leidel",- "file_count": 3,- "pct": 0.03571428571428571+ "name": "Adam Franke",+ "first_commit_at": "2026-04-03T07:20:45+00:00"},{- "name": "emisargent",- "file_count": 1,- "pct": 0.011904761904761904+ "name": "Lakshay Sachdeva",+ "first_commit_at": "2026-04-03T12:53:35+00:00"},{- "name": "Hendrik Makait",- "file_count": 1,- "pct": 0.011904761904761904+ "name": "Gabriel Mitelman Tkacz",+ "first_commit_at": "2026-04-04T09:56:16+00:00"}- ],- "single_owner_files": 28,- "silo_count": 4,- "truck_factor": 3+ ]},- "quality": {- "average_health": 6.94,- "maintainability_average": 7.71,- "performance_average": 9.96,- "worst_performer_path": "src/flask/app.py",- "worst_performer_score": 1.07,- "open_findings": 209,- "severity_breakdown": {- "critical": 8,- "high": 32,- "medium": 84,- "low": 85- },- "defect_accuracy": {- "k": 20,- "hits": 5,- "precision": 0.25,- "base_rate": 0.0556,- "lift": 4.5,- "window_days": 180,- "scored_files": 108,- "defect_files": 6,- "concentration_file_fraction": 0.2,- "concentration_defect_share": 0.8333,- "precision_table": [- {- "k": 10,- "hits": 3- },- {- "k": 20,- "hits": 5- },- {- "k": 30,- "hits": 5- }- ],- "flagged_files": [- {- "file_path": "src/flask/app.py",- "score": 1.07,- "recent_fixes": 1- },- {- "file_path": "src/flask/sansio/app.py",- "score": 1.35,- "recent_fixes": 1- },- {- "file_path": "src/flask/cli.py",- "score": 2.05,- "recent_fixes": 0- },- {- "file_path": "src/flask/sansio/blueprints.py",- "score": 3.53,- "recent_fixes": 0- },- {- "file_path": "src/flask/config.py",- "score": 5.15,- "recent_fixes": 0- },- {- "file_path": "src/flask/testing.py",- "score": 5.5,- "recent_fixes": 0- },- {- "file_path": "src/flask/ctx.py",- "score": 6.51,- "recent_fixes": 0- },- {- "file_path": "src/flask/templating.py",- "score": 7.15,- "recent_fixes": 0- },- {- "file_path": "src/flask/sansio/scaffold.py",- "score": 7.16,- "recent_fixes": 0- },- {- "file_path": "tests/test_basic.py",- "score": 7.49,- "recent_fixes": 1- },- {- "file_path": "src/flask/debughelpers.py",- "score": 7.53,- "recent_fixes": 0- },- {- "file_path": "src/flask/helpers.py",- "score": 8.09,- "recent_fixes": 0- },- {- "file_path": "tests/test_blueprints.py",- "score": 8.14,- "recent_fixes": 1- },- {- "file_path": "tests/test_cli.py",- "score": 8.19,- "recent_fixes": 1- },- {- "file_path": "src/flask/wrappers.py",- "score": 8.59,- "recent_fixes": 0- },- {- "file_path": "src/flask/blueprints.py",- "score": 8.64,- "recent_fixes": 0- },- {- "file_path": "tests/test_templating.py",- "score": 8.74,- "recent_fixes": 0- },- {- "file_path": "tests/test_appctx.py",- "score": 8.94,- "recent_fixes": 0- },- {- "file_path": "tests/test_reqctx.py",- "score": 8.94,- "recent_fixes": 0- },- {- "file_path": "tests/test_testing.py",- "score": 8.94,- "recent_fixes": 0- }- ]- },- "distribution": {- "total_files": 108,- "total_nloc": 14046,- "bands": {- "healthy": {- "files": 97,- "nloc": 7562,- "pct": 53.8- },- "warning": {- "files": 7,- "nloc": 3192,- "pct": 22.7- },- "alert": {- "files": 4,- "nloc": 3292,- "pct": 23.4- }- }- },- "doc_coverage_pct": 0,- "page_count": 0,- "test_coverage_pct": 62,- "dead_code": {- "total_findings": 28,- "deletable_lines": 169- }- },- "knowledge": {- "decision_count": 32,- "active_decision_count": 0- },- "superlatives": {+ "records": {"largest_file": {- "path": "tests/test_basic.py",- "nloc": 1440+ "path": "packages/api-client/src/__fixtures__/hosted/health-overview.json",+ "nloc": 1707+ },+ "gnarliest_file": {+ "path": "packages/server/src/repowise/server/mcp_server/tool_answer/answer.py",+ "max_ccn": 166},"most_complex_symbol": {- "name": "register",- "file_path": "src/flask/sansio/blueprints.py",- "complexity": 22+ "name": "get_answer",+ "file_path": "packages/server/src/repowise/server/mcp_server/tool_answer/answer.py",+ "complexity": 166},"most_changed_file": {- "path": "src/flask/app.py",- "commit_count": 134+ "path": "uv.lock",+ "commit_count": 65},"oldest_file": {- "path": "tests/test_apps/blueprintapp/apps/__init__.py",- "first_commit_at": "2014-09-11T20:09:50+00:00"+ "path": "package-lock.json",+ "first_commit_at": "2026-03-23T09:14:58+00:00"+ },+ "most_central_file": {+ "path": "packages/core/src/repowise/core/persistence/models.py",+ "pagerank": 0.0098,+ "import_count": 188},"strongest_coupling": {- "a": "CHANGES.rst",- "b": "src/flask/sansio/app.py",- "count": 3+ "a": "packages/cli/src/repowise/cli/__init__.py",+ "b": "pyproject.toml",+ "count": 42.51+ },+ "largest_cycle": {+ "files": 151,+ "cycle_count": 18+ },+ "symbol_shape": {+ "total": 24412,+ "async_count": 2882,+ "async_pct": 11.8,+ "documented_count": 6461,+ "documented_pct": 26.5+ },+ "biggest_commit": {+ "sha": "a901cc985cbd47634a45dc50ddab2d80f2c36cc4",+ "subject": "feat: v0.3.0 \u2014 workspaces, auto-sync hooks, graph intelligence (#76)",+ "lines_changed": 27519,+ "files_changed": 187+ },+ "widest_commit": {+ "sha": "aadd8f5e022208cc0901bd34d03bd6287aee43a6",+ "subject": "release: v0.5.0 \u2014 frontend monorepo restructure, build hardening, jobs reliability (#132)",+ "files_changed": 278,+ "lines_changed": 20715}- },- "graph": {- "cycle_clusters": 7,- "files_in_cycles": 63,- "largest_cycle": 21,- "community_count": 48- },- "build": {- "page_count": 486,- "total_tokens": 5231044,- "cost_usd": 0.92,- "llm_operations": 640}}diff --git a/packages/api-client/src/hosted.fixtures.test.ts b/packages/api-client/src/hosted.fixtures.test.tsindex 4a5feb1c0..e11885bfd 100644--- a/packages/api-client/src/hosted.fixtures.test.ts+++ b/packages/api-client/src/hosted.fixtures.test.ts@@ -123,9 +123,22 @@ describe("hosted fixtures satisfy the OSS contracts", () => {expect(summary.health).toBeDefined();expect(summary.sync).toBeDefined();+ // Assert every top-level section, not just two. `getStatsHighlights` is a+ // bare passthrough with no field mapping, so this fixture is the only thing+ // standing between a hosted payload drifting from the contract and the+ // Stats page rendering `undefined` sections at runtime.const stats: StatsHighlights = await p.getStatsHighlights("repo-1");- expect(stats.scale).toBeDefined();- expect(stats.people).toBeDefined();+ expect(stats.scale.size_class.name).toBeTruthy();+ expect(stats.origin).toBeDefined();+ expect(stats.rhythm.punch_card.matrix).toHaveLength(7);+ expect(stats.rhythm.punch_card.matrix[0]).toHaveLength(24);+ expect(["author_local", "utc"]).toContain(stats.rhythm.punch_card.timezone_mode);+ expect(stats.people.arrivals).toBeInstanceOf(Array);+ expect(stats.people.chronotypes).toBeInstanceOf(Array);+ expect(stats.records).toBeDefined();+ // `churn` is legitimately null when the history was too deep to walk, so+ // the key must exist while the value may not.+ expect(stats).toHaveProperty("churn");});it("files index and file detail", async () => {diff --git a/packages/cli/src/repowise/cli/commands/update_cmd/command.py b/packages/cli/src/repowise/cli/commands/update_cmd/command.pyindex 7f9535770..8f5620c5c 100644--- a/packages/cli/src/repowise/cli/commands/update_cmd/command.py+++ b/packages/cli/src/repowise/cli/commands/update_cmd/command.py@@ -51,6 +51,7 @@ from .mode import _resolve_index_only_modefrom .persistence import (_persist_index_only_update,_run_full_health_rescore,+ heal_commit_offsets,stamp_head_commit,)from .reporting import (@@ -748,6 +749,9 @@ def run_update(# current here while the DB head_commit is still the last full index.if not dry_run:stamp_head_commit(repo_path, head)+ # A capture added after this repo was indexed would otherwise wait+ # for the next commit to land, which on a quiet repo is never.+ heal_commit_offsets(repo_path)_refresh_editor_stamp(repo_path, agents_md)# The index is current, so any pending marker a bailed update left# is by definition caught-up (or older); drop it here too rather@@ -903,6 +907,9 @@ def run_update(# Keep the DB freshness stamp in lockstep with state.json: the server's# /repos endpoint reads head_commit from the row, not the state file.stamp_head_commit(repo_path, head)+ # This path skips the git phase, so its offset backfill never runs.+ if not dry_run:+ heal_commit_offsets(repo_path)_refresh_editor_stamp(repo_path, agents_md)# We hold the lock and have advanced to head; drop any stale pending# marker a bailed update left behind.diff --git a/packages/cli/src/repowise/cli/commands/update_cmd/persistence.py b/packages/cli/src/repowise/cli/commands/update_cmd/persistence.pyindex 8b66387bc..9c79ef46d 100644--- a/packages/cli/src/repowise/cli/commands/update_cmd/persistence.py+++ b/packages/cli/src/repowise/cli/commands/update_cmd/persistence.py@@ -129,6 +129,53 @@ def stamp_head_commit(repo_path: Any, head: str | None) -> None:run_async(reconcile_repo_head_commit(Path(repo_path), head))+def heal_commit_offsets(repo_path: Any) -> None:+ """Fill ``git_commits.committed_offset_minutes`` on a quiet repo.++ The commit walk captures this offset, so any update that ingests commits+ backfills it on the way past. A repo whose HEAD hasn't moved never gets+ there — it takes the "already up to date" fast path — which would strand an+ index written before the column existed with a UTC-only punch card until+ someone re-indexed. Nobody should have to re-index for a new column, so the+ fast path calls this too.++ Cheap enough to run unconditionally: one indexed SELECT that returns nothing+ once the column is filled, and no git at all in that case. Best-effort — a+ failure here must never turn a clean no-op into an error.+ """+ root = Path(repo_path)+ if not (root / ".repowise" / "wiki.db").is_file():+ return++ async def _run() -> None:+ from repowise.core.ingestion.git_indexer import GitIndexer+ from repowise.core.persistence import (+ create_engine,+ create_session_factory,+ get_session,+ init_db,+ )+ from repowise.core.persistence.crud import get_repository_by_path+ from repowise.core.persistence.database import resolve_db_url+ from repowise.core.pipeline.incremental import reconcile_commit_offsets++ engine = create_engine(resolve_db_url(root))+ try:+ # Migrates first: on the fast path this may be the run that adds the+ # column the backfill is about to fill.+ await init_db(engine)+ async with get_session(create_session_factory(engine)) as session:+ repo = await get_repository_by_path(session, str(root))+ if repo is None:+ return+ await reconcile_commit_offsets(session, repo.id, GitIndexer(root))+ finally:+ await engine.dispose()++ with contextlib.suppress(Exception):+ run_async(_run())++def _persist_index_only_update(repo_path: Any,graph_builder: Any,diff --git a/packages/core/alembic/versions/0043_stats_birthday_and_churn.py b/packages/core/alembic/versions/0043_stats_birthday_and_churn.pynew file mode 100644index 000000000..36f2ce29f--- /dev/null+++ b/packages/core/alembic/versions/0043_stats_birthday_and_churn.py@@ -0,0 +1,65 @@+"""Stats page: first-commit subject, lifetime churn, and per-commit UTC offset.++Three captures the Stats page needs that the existing tables cannot answer.++``repositories.first_commit_subject`` — the root commit's headline, so the page+can render the repo's opening line rather than just its date. Rides along free+with the root commit already loaded for ``first_commit_at`` (0037).++``repositories.total_lines_added`` / ``total_lines_deleted`` — lifetime churn.+Summing ``git_commits`` cannot answer this: that table is bounded to the newest+N commits, so a repo with more history than the window would silently report a+windowed figure as a lifetime one. NULL when the history was too deep to walk.++``git_commits.committed_offset_minutes`` — minutes east of UTC at commit time.+``committed_at`` is stored as a UTC instant, which discards the author's local+wall-clock, so time-of-day analysis (the punch card, per-author peak hour) read+a 10pm commit in Mumbai as a mid-afternoon one. Backfilled on the next update+(``pipeline.incremental.reconcile_commit_offsets``) rather than requiring a+re-index; rows stay NULL and fall back to UTC until then.++All nullable so existing indexes keep working untouched.++Revision ID: 0043+Revises: 0042+Create Date: 2026-07-26+"""++from __future__ import annotations++from collections.abc import Sequence++import sqlalchemy as sa+from alembic import op++# revision identifiers+revision: str = "0043"+down_revision: str | None = "0042"+branch_labels: str | Sequence[str] | None = None+depends_on: str | Sequence[str] | None = None+++def upgrade() -> None:+ op.add_column(+ "repositories",+ sa.Column("first_commit_subject", sa.Text(), nullable=True),+ )+ op.add_column(+ "repositories",+ sa.Column("total_lines_added", sa.Integer(), nullable=True),+ )+ op.add_column(+ "repositories",+ sa.Column("total_lines_deleted", sa.Integer(), nullable=True),+ )+ op.add_column(+ "git_commits",+ sa.Column("committed_offset_minutes", sa.Integer(), nullable=True),+ )+++def downgrade() -> None:+ op.drop_column("git_commits", "committed_offset_minutes")+ op.drop_column("repositories", "total_lines_deleted")+ op.drop_column("repositories", "total_lines_added")+ op.drop_column("repositories", "first_commit_subject")diff --git a/packages/core/src/repowise/core/ingestion/git_commit_index.py b/packages/core/src/repowise/core/ingestion/git_commit_index.pyindex 4563d6ded..68ee2a0f7 100644--- a/packages/core/src/repowise/core/ingestion/git_commit_index.py+++ b/packages/core/src/repowise/core/ingestion/git_commit_index.py@@ -260,6 +260,7 @@ def load_commit_index("author_name": header["author_name"],"author_email": header["author_email"],"ts": header["ts"],+ "tz_offset_minutes": header.get("tz_offset_minutes"),"subject": header["subject"],"changes": commit_changes,"agent_name": prov.agent,diff --git a/packages/core/src/repowise/core/ingestion/git_indexer/agent_provenance.py b/packages/core/src/repowise/core/ingestion/git_indexer/agent_provenance.pyindex 2fca8a110..ee8952792 100644--- a/packages/core/src/repowise/core/ingestion/git_indexer/agent_provenance.py+++ b/packages/core/src/repowise/core/ingestion/git_indexer/agent_provenance.py@@ -66,6 +66,7 @@ import jsonimport refrom collections import Counterfrom dataclasses import dataclass+from functools import lru_cachefrom pathlib import Pathimport structlog@@ -77,6 +78,7 @@ __all__ = ["AgentProvenanceClassifier","AgentTraceIndex","_agent_from_git_ai_note",+ "agent_from_identity","classifier_from_repo_config",]@@ -660,6 +662,38 @@ class AgentProvenanceClassifier:return _HUMAN+@lru_cache(maxsize=1)+def _default_classifier() -> AgentProvenanceClassifier:+ """Shared registry-only classifier for identity lookups (no repo config)."""+ return AgentProvenanceClassifier()+++def agent_from_identity(name: str | None, email: str | None) -> str | None:+ """The agent behind an author identity, or ``None`` for a human.++ Identity only — deliberately ignores commit-message trailers, which mark a+ commit a *human* authored with agent assistance. Callers asking "is this+ contributor a person?" (the stats page's people sections) need the identity+ answer: a human using Claude Code is still a human, while a commit authored+ by ``[email protected]`` is not.++ Also matches an agent's plain display name, since a locally-configured+ ``user.name`` of "Claude" carries no vendor e-mail to recognise.+ """+ hit = _default_classifier()._identity_match(email or "")+ if hit:+ return hit[0]+ slug = (name or "").strip().lower()+ if not slug:+ return None+ for token, agent in _AGENT_ALIASES:+ # Whole-token match only: "Aiden" must not read as "aider", and a+ # surname like "Coplin" must not read as "copilot".+ if re.fullmatch(rf"{re.escape(token)}([-_ ]?(code|cli|agent|ai|bot))?", slug):+ return agent+ return None++def classifier_from_repo_config(repo_path: object) -> AgentProvenanceClassifier:"""Build a classifier with any repo-local pattern extensions.diff --git a/packages/core/src/repowise/core/ingestion/git_indexer/commit_rows.py b/packages/core/src/repowise/core/ingestion/git_indexer/commit_rows.pyindex d1522c046..cdbca277d 100644--- a/packages/core/src/repowise/core/ingestion/git_indexer/commit_rows.py+++ b/packages/core/src/repowise/core/ingestion/git_indexer/commit_rows.py@@ -109,6 +109,10 @@ def build_commit_rows(parsed_commits: list[dict]) -> list[dict]:"author_name": c.get("author_name", "") or "","author_email": c.get("author_email", "") or "","committed_at": _committed_at(c.get("ts", 0) or 0),+ # Minutes east of UTC at the moment of the commit. ``committed_at``+ # is a UTC instant, so this is the only way back to the author's+ # local wall-clock for time-of-day analysis. None on older rows.+ "committed_offset_minutes": c.get("tz_offset_minutes"),"subject": subject,"lines_added": feats.la,"lines_deleted": feats.ld,diff --git a/packages/core/src/repowise/core/ingestion/git_indexer/indexer.py b/packages/core/src/repowise/core/ingestion/git_indexer/indexer.pyindex 2d1dd1011..79e1119e3 100644--- a/packages/core/src/repowise/core/ingestion/git_indexer/indexer.py+++ b/packages/core/src/repowise/core/ingestion/git_indexer/indexer.py@@ -14,7 +14,7 @@ import jsonimport osimport threadingimport time-from collections.abc import Callable+from collections.abc import Callable, Sequencefrom pathlib import Pathfrom typing import Any@@ -32,13 +32,24 @@ from .co_change import compute_co_changes, compute_co_changes_and_entropyfrom .enrich import compute_percentilesfrom .file_history import DECAY_REFRESH_KEYS, index_filefrom .prior_defects import FixWalk, PriorDefects, collect_fix_commits, compute_prior_defects-from .records import GitIndexSummary, _CommitRec, _should_skip_index, capture_repo_totals+from .records import (+ GitIndexSummary,+ _CommitRec,+ _should_skip_index,+ _tz_offset_minutes,+ capture_repo_totals,+)from .tiers import GitIndexTierlogger = structlog.get_logger(__name__)__all__ = ["GitIndexer"]+# Shas per ``git log --no-walk`` call when backfilling commit offsets. Each sha+# is a 40-char argv entry, and Windows caps a command line near 32k characters —+# 200 leaves generous headroom while keeping the subprocess count low.+_OFFSET_LOOKUP_CHUNK = 200+class GitIndexer:"""Mines git history into the git_metadata table.@@ -654,6 +665,49 @@ class GitIndexer:# broken call than a genuinely empty history.return shas or None+ def capture_commit_offsets(self, shas: Sequence[str]) -> dict[str, int]:+ """Map each requested sha to its commit's UTC offset in minutes.++ Backfill support for :attr:`GitCommit.committed_offset_minutes`, which+ older indexes predate.++ Looks the shas up directly (``--no-walk``) rather than walking HEAD and+ filtering: the persisted commit table can hold more rows than the+ indexer's current ``commit_limit`` (the window has moved, and co-change+ walks deeper), so a bounded walk silently leaves the oldest rows unfilled+ — which is exactly the half-filled state this backfill exists to avoid.+ Requests are chunked because every sha is an argv entry and Windows caps+ a command line at ~32k characters.++ Returns ``{}`` on total git failure; a failed chunk drops only its own+ shas. An unfilled offset degrades that row to UTC, which is what it+ already was.+ """+ if not shas:+ return {}+ repo = self._get_repo()+ if repo is None:+ return {}++ offsets: dict[str, int] = {}+ try:+ for start in range(0, len(shas), _OFFSET_LOOKUP_CHUNK):+ chunk = list(shas[start : start + _OFFSET_LOOKUP_CHUNK])+ try:+ out = repo.git.log("--no-walk", "--format=%H %cI", *chunk)+ except Exception as exc:+ logger.debug("commit_offsets_chunk_failed", error=str(exc))+ continue+ for line in out.split("\n"):+ sha, _, iso = line.strip().partition(" ")+ minutes = _tz_offset_minutes(iso)+ if sha and minutes is not None:+ offsets[sha] = minutes+ finally:+ with contextlib.suppress(Exception):+ repo.close()+ return offsets+def capture_new_fix_events(self, *, known_shas: set[str] | None = None) -> tuple[list[dict], int, set[str]]:diff --git a/packages/core/src/repowise/core/ingestion/git_indexer/records.py b/packages/core/src/repowise/core/ingestion/git_indexer/records.pyindex 592531734..e3206bb21 100644--- a/packages/core/src/repowise/core/ingestion/git_indexer/records.py+++ b/packages/core/src/repowise/core/ingestion/git_indexer/records.py@@ -4,7 +4,7 @@ from __future__ import annotationsimport refrom dataclasses import dataclass, field-from datetime import datetime+from datetime import UTC, datetimefrom pathlib import Pathfrom typing import Any@@ -30,6 +30,10 @@ __all__ = [_RECORD_SEP = "\x00"_FIELD_SEP = "\x1f"+# Same cap the per-commit row builder applies to ``subject`` (commit_rows), kept+# local so this module stays import-free of its sibling.+_MAX_SUBJECT_LEN = 500+# Shared log format for both the batched repo-wide index and the per-file# ``--follow`` fallback. NOTE: the format ARG uses git's literal ``%x00`` /# ``%x1f`` escape text — NOT the real bytes — because a real NUL in a subprocess@@ -45,7 +49,11 @@ _FIELD_SEP = "\x1f"# across several contributor buckets. Committer name/email (``%cN``/``%cE``)# ride the same record for agent-provenance classification: a service identity# as committer over a human author means an agent pushed/amended the change.-_LOG_FORMAT = "%x00%H%x1f%aN%x1f%aE%x1f%cN%x1f%cE%x1f%ct%x1f%P%x1f%s%x1f%b"+# ``%cI`` follows ``%ct`` purely for its UTC offset: ``%ct`` is a bare epoch, so+# the author's local wall-clock is unrecoverable from it. Time-of-day features+# (the stats punch card, per-author peak hour) are meaningless across timezones+# without it — a 10pm commit in Mumbai reads as mid-afternoon in UTC.+_LOG_FORMAT = "%x00%H%x1f%aN%x1f%aE%x1f%cN%x1f%cE%x1f%ct%x1f%cI%x1f%P%x1f%s%x1f%b"# A git ``--numstat`` line: ``<added>\t<deleted>\t<path>`` where added/deleted# are decimal counts or ``-`` for binary files. Used to split the trailing@@ -72,12 +80,33 @@ class _CommitRec:agent_tier: int | None = None+def _tz_offset_minutes(iso: str) -> int | None:+ """Minutes east of UTC for a git ``%cI`` timestamp, or ``None``.++ Only the offset is kept — the instant itself already rides on ``%ct``. Git+ always emits a numeric offset (never a name), so the tail is ``±HH:MM``.+ Parsed by hand rather than via ``fromisoformat`` because this runs once per+ commit on every index and the string shape is fixed.+ """+ iso = iso.strip()+ if len(iso) < 6:+ return None+ sign, hh, mm = iso[-6], iso[-5:-3], iso[-2:]+ if sign not in "+-" or iso[-3] != ":":+ return None+ try:+ minutes = int(hh) * 60 + int(mm)+ except ValueError:+ return None+ return -minutes if sign == "-" else minutes++def _parse_commit_record(record: str) -> tuple[dict, list[str]] | None:"""Parse one ``\\x00``-delimited git-log record into header + numstat lines.A record (after the leading record separator is stripped) looks like::- <sha>\\x1f<an>\\x1f<ae>\\x1f<cn>\\x1f<ce>\\x1f<ct>\\x1f<parents>\\x1f<subject>\\x1f<body>\\n<numstat...>+ <sha>\\x1f<an>\\x1f<ae>\\x1f<cn>\\x1f<ce>\\x1f<ct>\\x1f<cI>\\x1f<parents>\\x1f<subject>\\x1f<body>\\n<numstat...>The body is multi-line, so the trailing ``--numstat`` block is identifiedby the first line matching :data:`_NUMSTAT_RE`; everything before it is@@ -87,12 +116,12 @@ def _parse_commit_record(record: str) -> tuple[dict, list[str]] | None:``parents`` (the commit's parent shas)."""parts = record.split(_FIELD_SEP)- if len(parts) < 9:+ if len(parts) < 10:return None- sha, an, ae, cn, ce, ct, parents, subject = parts[:8]+ sha, an, ae, cn, ce, ct, ciso, parents, subject = parts[:9]# %b cannot contain the field separator, so the body + numstat tail is the- # 9th field exactly (join is defensive against an unexpected stray sep).- body_and_numstat = _FIELD_SEP.join(parts[8:])+ # 10th field exactly (join is defensive against an unexpected stray sep).+ body_and_numstat = _FIELD_SEP.join(parts[9:])body_lines: list[str] = []numstat_lines: list[str] = []@@ -118,6 +147,7 @@ def _parse_commit_record(record: str) -> tuple[dict, list[str]] | None:"committer_name": cn or "","committer_email": ce or "","ts": ts,+ "tz_offset_minutes": _tz_offset_minutes(ciso),"is_merge": len(parent_shas) > 1,# Parent shas ride along for the agent-trace channel, whose records# capture the pre-commit HEAD (i.e. a parent of the commit that@@ -200,23 +230,66 @@ class RepoTotals:first_commit_at: datetime | None = Nonetotal_contributor_count: int | None = Nonefirst_commit_author: str | None = None+ first_commit_subject: str | None = None+ total_lines_added: int | None = None+ total_lines_deleted: int | None = None+++# Lifetime churn walks every commit's shortstat, so unlike the other totals it+# is O(history). Above this many commits the walk is skipped and the fields stay+# None (the stats page hides the ledger rather than showing a windowed number+# as if it were lifetime). Raise it if a big repo ever wants the figure enough+# to pay for the walk.+_CHURN_COMMIT_CEILING = 50_000++# ``git log --shortstat`` summary line, e.g.+# " 3 files changed, 41 insertions(+), 9 deletions(-)". Either half can be absent.+_SHORTSTAT_RE = re.compile(r"(\d+) insertions?\(\+\)|(\d+) deletions?\(-\)")+++def _lifetime_churn(repo: Any, commit_count: int | None) -> tuple[int | None, int | None]:+ """Total lines ever added / deleted across the whole history.++ One ``git log --shortstat`` pass, summed here. The per-commit ``git_commits``+ table cannot answer this: it is bounded to the newest N commits, so summing+ it silently understates a repo with more history than the window.++ Skipped above :data:`_CHURN_COMMIT_CEILING` commits — returns ``(None, None)``+ so the caller stores nothing and the UI omits the figure.+ """+ if commit_count is None or commit_count > _CHURN_COMMIT_CEILING:+ return None, None+ try:+ out = repo.git.log("--shortstat", "--pretty=tformat:", "HEAD")+ except Exception:+ return None, None+ added = deleted = 0+ for ins, dels in _SHORTSTAT_RE.findall(out):+ if ins:+ added += int(ins)+ if dels:+ deleted += int(dels)+ return added, deleteddef capture_repo_totals(repo: Any) -> RepoTotals:"""Whole-history stats for *repo* via a handful of cheap git calls.- All three git calls are O(1) in subprocess count and independent of the- indexer's ``commit_limit``, so they stay cheap no matter how deep the+ The first three git calls are O(1) in subprocess count and independent of+ the indexer's ``commit_limit``, so they stay cheap no matter how deep thehistory is:- ``git rev-list --count HEAD`` — the true total commit count.- - the root commit(s) — earliest committed date (project age) and the- founding author's name. Multiple roots (merged histories) use the- earliest root.+ - the root commit(s) — earliest committed date (project age), the founding+ author's name, and that commit's subject. Multiple roots (merged+ histories) use the earliest root.- ``git shortlog -sn HEAD`` — one line per mailmap-folded author, so itsline count is the true all-time contributor count. Passing ``HEAD``keeps shortlog from blocking on stdin.+ Lifetime churn is the one exception: it walks the history, so it is capped+ (see :func:`_lifetime_churn`).+Each field is captured under its own ``try`` so one failure (empty/unbornHEAD, detached state, git unavailable) never voids the others."""@@ -232,8 +305,17 @@ def capture_repo_totals(repo: Any) -> RepoTotals:root_commits = [repo.commit(sha) for sha in roots if sha]if root_commits:oldest = min(root_commits, key=lambda c: c.committed_datetime)- totals.first_commit_at = oldest.committed_datetime+ # Normalise to UTC before storing. GitPython hands back the author's+ # local offset, and SQLite drops tzinfo on write — so the raw value+ # came back looking like a UTC instant while actually being local+ # wall-clock, shifting the founding date by the author's offset (and+ # across midnight for anyone far enough east or west).+ totals.first_commit_at = oldest.committed_datetime.astimezone(UTC)totals.first_commit_author = (oldest.author.name or None) if oldest.author else None+ # ``summary`` is the subject line; cap it like every other stored+ # subject so one pathological commit can't bloat the repo row.+ subject = (oldest.summary or "").strip() if oldest.summary else ""+ totals.first_commit_subject = subject[:_MAX_SUBJECT_LEN] or Noneexcept Exception:pass@@ -243,6 +325,10 @@ def capture_repo_totals(repo: Any) -> RepoTotals:except Exception:pass+ totals.total_lines_added, totals.total_lines_deleted = _lifetime_churn(+ repo, totals.total_commit_count+ )+return totalsdiff --git a/packages/core/src/repowise/core/persistence/crud/analysis/health.py b/packages/core/src/repowise/core/persistence/crud/analysis/health.pyindex 2bed2a4f7..47ada9c1f 100644--- a/packages/core/src/repowise/core/persistence/crud/analysis/health.py+++ b/packages/core/src/repowise/core/persistence/crud/analysis/health.py@@ -347,9 +347,18 @@ async def get_perf_coverage(session: AsyncSession, repository_id: str) -> PerfCoreturn coverage_for_metrics(metrics, lang_by_path)-async def get_health_summary(session: AsyncSession, repository_id: str) -> dict:- """Aggregate KPIs over the per-file metrics table."""- metrics = await get_health_metrics(session, repository_id)+async def get_health_summary(+ session: AsyncSession, repository_id: str, *, metrics: list | None = None+) -> dict:+ """Aggregate KPIs over the per-file metrics table.++ *metrics* lets a caller that has already loaded the per-file rows hand them+ over instead of paying for a second full read. Callers doing more than one+ thing with them — the overview payload computes both these KPIs and the+ defect-accuracy stat — would otherwise pull the whole table twice.+ """+ if metrics is None:+ metrics = await get_health_metrics(session, repository_id)if not metrics:return {"file_count": 0,diff --git a/packages/core/src/repowise/core/persistence/crud/git.py b/packages/core/src/repowise/core/persistence/crud/git.pyindex 701a73877..2a976cdfa 100644--- a/packages/core/src/repowise/core/persistence/crud/git.py+++ b/packages/core/src/repowise/core/persistence/crud/git.py@@ -265,6 +265,28 @@ async def upsert_git_commits_bulk()+async def get_commits_missing_offset(+ session: AsyncSession, repository_id: str, *, limit: int = 20_000+) -> list[str]:+ """Shas whose ``committed_offset_minutes`` was never captured.++ Indexes written before the offset existed have it NULL on every row, which+ would leave the punch card mixing author-local and UTC hours. The update+ path backfills them (see ``pipeline.incremental.reconcile_commit_offsets``);+ this is the "what still needs filling" half. Bounded so one update can't+ balloon on a very deep window — the next update picks up the remainder.+ """+ rows = await session.execute(+ select(GitCommit.sha)+ .where(+ GitCommit.repository_id == repository_id,+ GitCommit.committed_offset_minutes.is_(None),+ )+ .limit(limit)+ )+ return [sha for (sha,) in rows]++async def delete_git_commits(session: AsyncSession, repository_id: str) -> None:"""Remove all per-commit rows for a repository (used before a clean reindex)."""await session.execute(delete(GitCommit).where(GitCommit.repository_id == repository_id))diff --git a/packages/core/src/repowise/core/persistence/crud/repository.py b/packages/core/src/repowise/core/persistence/crud/repository.pyindex 67fb52411..361024ff3 100644--- a/packages/core/src/repowise/core/persistence/crud/repository.py+++ b/packages/core/src/repowise/core/persistence/crud/repository.py@@ -136,6 +136,9 @@ async def update_repo_git_totals(first_commit_at: datetime | None = None,total_contributor_count: int | None = None,first_commit_author: str | None = None,+ first_commit_subject: str | None = None,+ total_lines_added: int | None = None,+ total_lines_deleted: int | None = None,) -> None:"""Store a repo's whole-history git totals, captured at index time (#730).@@ -150,6 +153,9 @@ async def update_repo_git_totals("first_commit_at": first_commit_at,"total_contributor_count": total_contributor_count,"first_commit_author": first_commit_author,+ "first_commit_subject": first_commit_subject,+ "total_lines_added": total_lines_added,+ "total_lines_deleted": total_lines_deleted,}if all(v is None for v in updates.values()):returndiff --git a/packages/core/src/repowise/core/persistence/models.py b/packages/core/src/repowise/core/persistence/models.pyindex 320f8c592..40a5d8e58 100644--- a/packages/core/src/repowise/core/persistence/models.py+++ b/packages/core/src/repowise/core/persistence/models.py@@ -63,6 +63,13 @@ class Repository(Base):# first-commit date). Both NULL until the first index writes them.total_contributor_count: Mapped[int | None] = mapped_column(Integer, nullable=True)first_commit_author: Mapped[str | None] = mapped_column(String(255), nullable=True)+ # The root commit's subject line, and lifetime churn across the whole+ # history. Both share the #730 reasoning: the bounded ``git_commits`` sample+ # cannot answer either without understating a long-lived repo. Churn is NULL+ # when the history was too deep to walk (see ``_lifetime_churn``).+ first_commit_subject: Mapped[str | None] = mapped_column(Text, nullable=True)+ total_lines_added: Mapped[int | None] = mapped_column(Integer, nullable=True)+ total_lines_deleted: Mapped[int | None] = mapped_column(Integer, nullable=True)settings_json: Mapped[str] = mapped_column(Text, nullable=False, default="{}")created_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), nullable=False, default=_now_utc@@ -555,6 +562,12 @@ class GitCommit(Base):author_name: Mapped[str] = mapped_column(String(255), nullable=False, default="")author_email: Mapped[str] = mapped_column(String(255), nullable=False, default="")committed_at: Mapped[datetime | None] = mapped_column(DateTime(timezone=True), nullable=True)+ # Minutes east of UTC when the commit was made. ``committed_at`` is stored as+ # a UTC instant, which loses the author's local wall-clock — so time-of-day+ # analysis (stats punch card, per-author peak hour) needs this to avoid+ # reporting a 10pm commit in Mumbai as a mid-afternoon one. NULL on rows+ # written before the offset was captured.+ committed_offset_minutes: Mapped[int | None] = mapped_column(Integer, nullable=True)subject: Mapped[str] = mapped_column(Text, nullable=False, default="")# Kamei change features (diff size + diffusion of THIS change)diff --git a/packages/core/src/repowise/core/pipeline/incremental.py b/packages/core/src/repowise/core/pipeline/incremental.pyindex 93936a830..7f1da02f6 100644--- a/packages/core/src/repowise/core/pipeline/incremental.py+++ b/packages/core/src/repowise/core/pipeline/incremental.py@@ -583,6 +583,9 @@ async def persist_incremental_commits(session: Any, repo_id: str, repo_path: Anyawait upsert_git_commits_bulk(session, repo_id, rows)await reconcile_commit_experience(session, repo_id, indexer)+ # Fills the commit-offset column on indexes written before it existed, so a+ # new capture never needs a re-index to become useful.+ await reconcile_commit_offsets(session, repo_id, indexer)# Refresh the repo-level whole-history totals so age / commit / contributor# counts keep growing between full re-indexes (#730). Cheap git calls, and@@ -595,6 +598,9 @@ async def persist_incremental_commits(session: Any, repo_id: str, repo_path: Anyfirst_commit_at=totals.first_commit_at,total_contributor_count=totals.total_contributor_count,first_commit_author=totals.first_commit_author,+ first_commit_subject=totals.first_commit_subject,+ total_lines_added=totals.total_lines_added,+ total_lines_deleted=totals.total_lines_deleted,)await persist_incremental_fix_events(session, repo_id, indexer)@@ -654,6 +660,45 @@ async def reconcile_commit_experience(session: Any, repo_id: str, indexer: Any)logger.debug("commit_experience_reconcile_failed", error=str(exc))+async def reconcile_commit_offsets(session: Any, repo_id: str, indexer: Any) -> None:+ """Backfill ``committed_offset_minutes`` on commits indexed before it existed.++ The offset is captured on the commit walk, so a full re-index gets it for+ free — but an update only writes rows for *new* commits, which would leave a+ repo indexed earlier with author-local hours on recent commits and UTC on+ everything older. A punch card mixing the two is worse than one that is+ honestly all-UTC, so the gap is closed here rather than waiting for a+ re-index nobody should have to run for a new column.++ One ``git log`` and one bulk update, both skipped entirely once the column+ is filled — the common case is a single cheap SELECT that returns nothing.+ Failure-isolated like the other update-time reconciles; an unfilled offset+ just means those rows keep falling back to UTC.+ """+ from repowise.core.persistence.crud import (+ get_commits_missing_offset,+ upsert_git_commits_bulk,+ )++ try:+ pending = await get_commits_missing_offset(session, repo_id)+ if not pending:+ return++ offsets = await asyncio.to_thread(indexer.capture_commit_offsets, pending)+ if not offsets:+ return++ await upsert_git_commits_bulk(+ session,+ repo_id,+ [{"sha": sha, "committed_offset_minutes": minutes} for sha, minutes in offsets.items()],+ )+ logger.info("commit_offsets_backfilled", repo_id=repo_id, count=len(offsets))+ except Exception as exc:+ logger.debug("commit_offset_reconcile_failed", error=str(exc))++def _committed_ts(committed_at: Any) -> float:"""``committed_at`` as a sortable epoch, or 0 when the row has no date.diff --git a/packages/core/src/repowise/core/pipeline/persist.py b/packages/core/src/repowise/core/pipeline/persist.pyindex af79c25f3..21f028754 100644--- a/packages/core/src/repowise/core/pipeline/persist.py+++ b/packages/core/src/repowise/core/pipeline/persist.py@@ -865,6 +865,9 @@ async def persist_git(result: Any, session: Any, repo_id: str) -> None:first_commit_at=totals.first_commit_at,total_contributor_count=totals.total_contributor_count,first_commit_author=totals.first_commit_author,+ first_commit_subject=totals.first_commit_subject,+ total_lines_added=totals.total_lines_added,+ total_lines_deleted=totals.total_lines_deleted,)diff --git a/packages/server/src/repowise/server/routers/overview.py b/packages/server/src/repowise/server/routers/overview.pyindex 1b92592b7..10aede542 100644--- a/packages/server/src/repowise/server/routers/overview.py+++ b/packages/server/src/repowise/server/routers/overview.py@@ -278,7 +278,11 @@ async def overview_summary()# --- Health KPIs + deltas vs previous snapshot ------------------------- health_summary = await crud.get_health_summary(session, repo_id)+ # Loaded once and handed to both consumers below: the KPI rollup and the+ # defect-accuracy stat both want every per-file row, and letting each fetch+ # its own pulled the whole table twice per page load.+ health_metrics = await crud.get_health_metrics(session, repo_id)+ health_summary = await crud.get_health_summary(session, repo_id, metrics=health_metrics)snapshots = await crud.list_health_snapshots(session, repo_id)hotspot_health: float | None = Nonelast_indexed_at: str | None = None@@ -310,6 +314,38 @@ async def overview_summary(if s in severity_breakdown:severity_breakdown[s] += 1+ # "Can you trust this score?" — the backtested precision of the defect+ # ranking, shown on the health card. Sourced here rather than from the stats+ # payload: this is a health number, and having the overview reach across+ # into another page's payload for it is what broke when that payload changed.+ defect_accuracy = None+ try:+ from repowise.core.analysis.health.defect_accuracy import compute_defect_accuracy++ defect_accuracy = compute_defect_accuracy(+ [+ {+ "file_path": m.file_path,+ "score": m.score,+ "nloc": m.nloc,+ "has_test_file": m.has_test_file,+ "module": m.module,+ }+ for m in health_metrics+ ],+ [+ {+ "file_path": f.file_path,+ "biomarker_type": f.biomarker_type,+ "severity": f.severity,+ }+ for f in findings+ ],+ )+ except Exception:+ # Best-effort: the card omits the panel rather than failing the page.+ defect_accuracy = None+# --- Attention items + onboarding targets -----------------------------decision_health = await crud.get_decision_health_summary(session, repo_id)knowledge = await compute_knowledge_map(session, repo_id)@@ -437,6 +473,7 @@ async def overview_summary("worst_performance_path": health_summary.get("worst_performance_path"),"worst_performance_score": health_summary.get("worst_performance_score"),"severity_breakdown": severity_breakdown,+ "defect_accuracy": defect_accuracy,"last_indexed_at": last_indexed_at,"snapshot_count": len(snapshots),"history": [diff --git a/packages/server/src/repowise/server/routers/stats.py b/packages/server/src/repowise/server/routers/stats.pyindex 833581a1c..048b768ff 100644--- a/packages/server/src/repowise/server/routers/stats.py+++ b/packages/server/src/repowise/server/routers/stats.py@@ -1,40 +1,42 @@"""/api/repos/{repo_id}/stats/highlights — the "By the Numbers" payload.-A single read-only aggregate that powers the repo Stats page: a showcase of-signals the engine already computes across every layer (graph, git, health,-docs, decisions, dead code) plus a handful of fun superlatives and a derived-"size class" label. Nothing here is new analysis — it stitches together rows-the indexer already wrote, so the page is cheap and never blocks.--Every section is built defensively: a missing table or column degrades that-section to ``None`` / empty rather than 500-ing the whole page, mirroring the-overview-summary contract.+A single read-only aggregate powering the repo Stats page. Its scope is defined+by subtraction: **only signals no other page in the app already shows**. Health+scores live on Code Health, commit volume and categories on Commits, per-person+ownership on Contributors, dependencies and communities on Architecture. What is+left, and what this endpoint serves, is the repo's identity (scale, origin,+lifetime churn), its rhythm (time-of-day, streaks, how fast code goes cold), and+its records — none of which have a home anywhere else.++Everything here reads rows the indexer already wrote; nothing recomputes+analysis. The three source tables are each pulled once, as narrow column+selects rather than whole ORM rows, and every derived figure rides one of those+three passes. Sections build defensively: a missing table degrades that section+to ``None`` / empty rather than 500-ing the page."""from __future__ import annotationsimport jsonimport re-from datetime import timedelta+from datetime import UTC, datetime, timedeltafrom itertools import pairwisefrom typing import Anyfrom fastapi import APIRouter, Depends, HTTPException-from sqlalchemy import func, select+from sqlalchemy import Integer, func, selectfrom sqlalchemy.ext.asyncio import AsyncSessionfrom repowise.core.ingestion.git_indexer import build_identity_resolver+from repowise.core.ingestion.git_indexer.agent_provenance import agent_from_identityfrom repowise.core.persistence import crudfrom repowise.core.persistence.models import (- DecisionRecord,- ExternalSystem,GitCommit,GitMetadata,GraphMetric,GraphNode,GraphNodeMembership,- LlmCost,- Page,+ HealthFileMetric,WikiSymbol,)from repowise.server.deps import get_db_session, verify_api_key@@ -62,6 +64,63 @@ _SIZE_CLASSES: tuple[tuple[int, str, str], ...] = ()_MEGALOPOLIS = ("Megalopolis", "A vast megalopolis — its own self-contained world.")+# Share of dated commits that must carry a UTC offset before the punch card and+# the chronotype awards switch from UTC to author-local time. Below it the page+# stays honestly all-UTC rather than mixing two clocks in one matrix — a chart+# where some cells are local and some are not is worse than one that is+# uniformly shifted. An index written before the offset column existed starts at+# 0% and fills on `repowise update` (the backfill in+# ``pipeline.incremental.reconcile_commit_offsets``), so this flips on its own.+# The backfill is capped per run, so a very deep window may need a few updates.+_LOCAL_TIME_COVERAGE = 0.99++# Minimum commits before someone is eligible for a chronotype award — a peak+# hour drawn from three commits is noise, not a habit.+_CHRONOTYPE_MIN_COMMITS = 10++# Automation, excluded from the people-shaped sections. A bot's "peak hour" is+# a cron schedule and its "arrival" is the day someone enabled it — both are+# facts about configuration, not about anyone's working habits, and leaving+# dependabot in a night-owl leaderboard makes the whole list read as noise.+# Matched on the author identity rather than on `agent_name`, which marks+# agent-*assisted* commits a human still authored.+# Either an explicit bot marker, or a service name matched in FULL. Matching a+# leading token would read "Netlify Johnson" as automation, and dropping a real+# person off the page is far worse than keeping one stray bot on it.+_BOT_NAME_RE = re.compile(+ r"(\[bot\]"+ r"|^bot$"+ r"|[-_ ]bot$"+ r"|^(dependabot|renovate(bot)?|greenkeeper|snyk([-_ ]bot)?|imgbot|"+ r"github[-_ ]?actions|semantic[-_ ]release|allcontributors|codecov|mergify|"+ r"pre[-_ ]commit[-_ ]ci|netlify|vercel)$)",+ re.IGNORECASE,+)+_BOT_EMAIL_RE = re.compile(+ r"(\[bot\]@|@bots\.noreply\.github\.com|^(actions@github\.com|"+ r"noreply@github\.com)$)",+ re.IGNORECASE,+)+++def _is_bot(name: str | None, email: str | None) -> bool:+ """True when this author is automation rather than a person.++ Two sources. Coding agents come from the ingestion layer's provenance+ registry, so "is this Claude / Cursor / Codex?" stays answered in exactly+ one place and the stats page inherits every vendor identity the indexer+ already knows. The regexes above cover the rest of CI automation, which+ provenance has no reason to model.++ Identity only: an agent-*assisted* commit still has a human author, and+ that person belongs in these lists.+ """+ if agent_from_identity(name, email):+ return True+ if name and _BOT_NAME_RE.search(name):+ return True+ return bool(email and _BOT_EMAIL_RE.search(email))+def _size_class(total_nloc: int) -> dict[str, Any]:for ceiling, name, blurb in _SIZE_CLASSES:@@ -75,6 +134,21 @@ def _iso(dt: Any) -> str | None:return dt.isoformat() if dt is not None else None+def _as_utc(dt: datetime | None) -> datetime | None:+ """Reinterpret a naive datetime as UTC.++ The ``committed_at`` columns are declared ``DateTime(timezone=True)``, but+ SQLite does not persist tzinfo, so a value read back is naive even though it+ was written aware. Anything doing arithmetic or calendar work on it has to+ restore the tz first or it silently mixes aware and naive values (and+ ``.hour`` quietly answers for the wrong clock). Same guard the change-risk+ tool applies for the same reason.+ """+ if dt is None:+ return None+ return dt if dt.tzinfo is not None else dt.replace(tzinfo=UTC)++# Test-path convention shared with analysis (communities / execution_flows):# catches conftest, fixtures, and spec files the `is_test` flag misses._TEST_PATH_RE = re.compile(@@ -82,6 +156,11 @@ _TEST_PATH_RE = re.compile()+# ---------------------------------------------------------------------------+# Scale — what this codebase is, dimensionally.+# ---------------------------------------------------------------------------++async def _scale(session: AsyncSession, repo_id: str, metrics: list[Any]) -> dict[str, Any]:"""Graph + NLOC scale signals + the size-class label."""file_count = (@@ -98,15 +177,6 @@ async def _scale(session: AsyncSession, repo_id: str, metrics: list[Any]) -> dic)or 0)- entry_point_count = (- await session.scalar(- select(func.count(GraphNode.id)).where(- GraphNode.repository_id == repo_id,- GraphNode.is_entry_point.is_(True),- )- )- or 0- )total_nloc = sum(int(m.nloc or 0) for m in metrics)lang_rows = await session.execute(@@ -123,7 +193,6 @@ async def _scale(session: AsyncSession, repo_id: str, metrics: list[Any]) -> dicreturn {"file_count": file_count,"symbol_count": symbol_count,- "entry_point_count": entry_point_count,"module_count": module_count,"total_nloc": total_nloc,"language_count": len(languages),@@ -132,14 +201,22 @@ async def _scale(session: AsyncSession, repo_id: str, metrics: list[Any]) -> dic}-def _punch_card_summary(punch: list[list[int]], dated_total: int) -> dict[str, Any]:+# ---------------------------------------------------------------------------+# Rhythm — the time-shape of the work. Nothing else in the app has a clock.+# ---------------------------------------------------------------------------+++def _punch_card_summary(+ punch: list[list[int]], dated_total: int, *, timezone_mode: str+) -> dict[str, Any]:"""Fold the weekday x hour commit matrix into a renderable summary.- ``matrix`` is 7 rows (0=Monday) x 24 hours in the stored UTC. Also names the- single hottest cell and the busiest weekday / peak hour (by marginal- totals) — the human-readable hooks the hero renders. The weekend share is- deliberately not computed here: which days are the weekend is a reader- preference, and the matrix already carries every weekday total."""+ ``matrix`` is 7 rows (0=Monday) x 24 hours in *timezone_mode*'s clock. Also+ names the single hottest cell and the busiest weekday / peak hour (by+ marginal totals). The weekend share is deliberately not computed here: which+ days are the weekend is a reader preference, and the matrix already carries+ every weekday total.+ """peak = {"weekday": 0, "hour": 0, "count": 0}for wd in range(7):for hr in range(24):@@ -157,6 +234,7 @@ def _punch_card_summary(punch: list[list[int]], dated_total: int) -> dict[str, A"busiest_weekday": busiest_weekday,"peak_hour": peak_hour,"total": dated_total,+ "timezone_mode": timezone_mode,}@@ -167,7 +245,8 @@ def _commit_velocity(commit_times: list[Any], last_at: Any) -> dict[str, Any]:an index that hasn't been synced today: it compares the 90 days ending atthe latest commit against the 90 before that. ``pct_change`` is None when theprior window is empty (a young repo), so the UI can omit a divide-by-zero- arrow rather than show a fake spike."""+ arrow rather than show a fake spike.+ """if last_at is None or not commit_times:return {"recent_90d": 0, "prior_90d": 0, "pct_change": None}recent_cut = last_at - timedelta(days=90)@@ -178,175 +257,311 @@ def _commit_velocity(commit_times: list[Any], last_at: Any) -> dict[str, Any]:return {"recent_90d": recent, "prior_90d": prior, "pct_change": pct_change}-async def _activity(session: AsyncSession, repo_id: str, repo: Any) -> dict[str, Any]:- """Commit volume, project age, agent-vs-human split, and a monthly series.+def _longest_streak(commit_days: set[Any]) -> dict[str, Any] | None:+ """Longest run of consecutive calendar days with at least one commit."""+ if not commit_days:+ return None+ days_sorted = sorted(commit_days)+ best_len = run_len = 1+ best_end = days_sorted[0]+ for prev, cur in pairwise(days_sorted):+ run_len = run_len + 1 if (cur - prev).days == 1 else 1+ if run_len > best_len:+ best_len, best_end = run_len, cur+ if best_len < 2:+ return None+ return {+ "days": best_len,+ "start": (best_end - timedelta(days=best_len - 1)).isoformat(),+ "end": best_end.isoformat(),+ }+++def _chronotypes(+ per_author_hours: dict[str, list[int]], names: dict[str, str]+) -> list[dict[str, Any]]:+ """Each frequent contributor's peak commit hour, with a habit label.++ Only meaningful once commits carry their author's UTC offset — the caller+ skips this entirely in UTC mode rather than award "night owl" to whoever+ happens to live furthest east.+ """+ out: list[dict[str, Any]] = []+ for key, hours in per_author_hours.items():+ total = sum(hours)+ if total < _CHRONOTYPE_MIN_COMMITS:+ continue+ peak_hour = max(range(24), key=lambda h: hours[h])+ # Off-hours share doubles as the tiebreaker for the label: a peak at 9am+ # with a third of commits after midnight is still a night owl.+ night = sum(hours[h] for h in (22, 23, 0, 1, 2, 3, 4))+ early = sum(hours[h] for h in (5, 6, 7, 8))+ if peak_hour >= 22 or peak_hour <= 4:+ label = "night_owl"+ elif 5 <= peak_hour <= 8:+ label = "early_bird"+ elif night / total >= 0.25:+ label = "night_owl"+ else:+ label = "daylight"+ out.append(+ {+ "name": names.get(key, key),+ "commits": total,+ "peak_hour": peak_hour,+ "label": label,+ "night_pct": round(night / total * 100.0, 1),+ "early_pct": round(early / total * 100.0, 1),+ }+ )+ return sorted(out, key=lambda a: -a["commits"])[:8]- Buckets the bounded ``git_commits`` table in Python so it is portable- across SQLite/Postgres date functions (same approach as the agent-trend- endpoint).++async def _commit_pass(session: AsyncSession, repo_id: str, repo: Any) -> dict[str, Any]:+ """Single scan of ``git_commits`` producing origin, rhythm and people signals.++ One query, one loop. Everything time-shaped the page shows — the punch card,+ streaks, the busiest day, the widest commit, momentum, chronotypes, arrival+ dates — falls out of this pass, because each of them needs per-commit rows+ and a second walk would double the cost for no new data.Headline totals (commit count, project age, contributor count, founder)- prefer the whole-history values captured on the ``Repository`` row at index- time — the ``git_commits`` table is bounded to the newest N commits, so- deriving them from it undercounts a long-lived repo badly (issue #730). The- bounded scan still drives the *sample* signals (agent/fix ratios, monthly- series, awards) that are meaningful on the recent window.-- Also derives the ``biggest_commit`` and ``longest_streak`` awards on the- same pass — they need per-commit rows anyway, and riding along here keeps- the endpoint at a single scan of the commits table. The caller moves them- into the superlatives payload."""+ prefer the whole-history values stamped on the ``Repository`` row at index+ time: ``git_commits`` is bounded to the newest N commits, so deriving them+ from it undercounts a long-lived repo badly (issue #730).+ """rows = (await session.execute(select(GitCommit.committed_at,+ GitCommit.committed_offset_minutes,GitCommit.author_name,GitCommit.author_email,- GitCommit.agent_name,- GitCommit.is_fix,GitCommit.sha,GitCommit.subject,GitCommit.lines_added,GitCommit.lines_deleted,GitCommit.files_changed,- GitCommit.change_risk_level,).where(GitCommit.repository_id == repo_id))).all()# Fold GitHub noreply variants and same-name real+noreply emails to one# identity so the same person isn't counted as several contributors.- resolve = build_identity_resolver([(name, email) for _, name, email, *_ in rows])+ resolve = build_identity_resolver([(name, email) for _, _, name, email, *_ in rows])total = 0- agent_total = 0- fix_total = 0- months: dict[str, dict[str, int]] = {}- agent_names: dict[str, int] = {}+ dated = 0+ with_offset = 0contributors: set[str] = set()- first_at: Any = None+ display_name: dict[str, str] = {}+ human_keys: set[str] = set()+ arrival: dict[str, datetime] = {}+ first_at: datetime | None = Nonefirst_sha: str | None = None- last_at: Any = None- commit_days: set[Any] = set()- # Coding-rhythm punch card: commits bucketed by weekday (0=Mon) x hour (UTC,- # the stored tz). Plus the raw timestamps for the recent-vs-prior velocity- # window and a low/moderate/high change-risk tally — all off this one scan.- punch = [[0] * 24 for _ in range(7)]- commit_times: list[Any] = []- risk_mix = {"low": 0, "moderate": 0, "high": 0}- # Top-2 commits by churn: the repo's very first commit is excluded from- # the "biggest commit" award (every import/initial commit would win), so- # two candidates are enough to survive dropping it.- top_commits: list[dict[str, Any]] = []-- for row in rows:- committed_at, author_name, author_email, agent_name, is_fix = row[:5]- sha, subject, lines_added, lines_deleted, files_changed, change_risk_level = row[5:]+ last_at: datetime | None = None+ commit_times: list[datetime] = []+ # Normalised (utc, offset) pairs kept for the second, timezone-aware pass:+ # whether the punch card is drawn in UTC or author-local can't be decided+ # until every row has been seen, so the calendar work waits for the verdict.+ stamps: list[tuple[datetime, int | None, str | None]] = []+ # Top-3 by churn and by breadth. The repo's very first commit is excluded+ # from both awards (an initial import would always win), so a couple of+ # spare candidates are enough to survive dropping it.+ top_churn: list[dict[str, Any]] = []+ top_wide: list[dict[str, Any]] = []++ for (+ committed_at,+ offset,+ author_name,+ author_email,+ sha,+ subject,+ added,+ deleted,+ files,+ ) in rows:total += 1+ key = Noneif author_email or author_name:key = resolve(author_name, author_email)if key:contributors.add(key)- if is_fix:- fix_total += 1- if change_risk_level in risk_mix:- risk_mix[change_risk_level] += 1- if agent_name:- agent_total += 1- agent_names[agent_name] = agent_names.get(agent_name, 0) + 1- churn = int(lines_added or 0) + int(lines_deleted or 0)+ display_name.setdefault(key, author_name or key)+ if not _is_bot(author_name, author_email):+ human_keys.add(key)++ churn = int(added or 0) + int(deleted or 0)if churn > 0:- top_commits.append(+ top_churn.append({"sha": sha,"subject": subject or "","lines_changed": churn,- "files_changed": int(files_changed or 0),+ "files_changed": int(files or 0),})- top_commits.sort(key=lambda c: -c["lines_changed"])- del top_commits[2:]- if committed_at is not None:- if first_at is None or committed_at < first_at:- first_at = committed_at- first_sha = sha- if last_at is None or committed_at > last_at:- last_at = committed_at- commit_days.add(committed_at.date())- commit_times.append(committed_at)- punch[committed_at.weekday()][committed_at.hour] += 1- key = committed_at.strftime("%Y-%m")- b = months.setdefault(key, {"total": 0, "agent": 0})- b["total"] += 1- if agent_name:- b["agent"] += 1-- biggest_commit = next((c for c in top_commits if c["sha"] != first_sha), None)-- longest_streak: dict[str, Any] | None = None- if commit_days:- days_sorted = sorted(commit_days)- best_len = run_len = 1- best_end = days_sorted[0]- for prev, cur in pairwise(days_sorted):- run_len = run_len + 1 if (cur - prev).days == 1 else 1- if run_len > best_len:- best_len, best_end = run_len, cur- if best_len >= 2:- longest_streak = {- "days": best_len,- "start": (best_end - timedelta(days=best_len - 1)).isoformat(),- "end": best_end.isoformat(),- }+ top_churn.sort(key=lambda c: -c["lines_changed"])+ del top_churn[3:]+ if int(files or 0) > 0:+ top_wide.append(+ {+ "sha": sha,+ "subject": subject or "",+ "files_changed": int(files or 0),+ "lines_changed": churn,+ }+ )+ top_wide.sort(key=lambda c: -c["files_changed"])+ del top_wide[3:]- monthly = [- {"month": m, "total": b["total"], "agent": b["agent"]} for m, b in sorted(months.items())- ]- busiest = max(monthly, key=lambda r: r["total"], default=None)+ moment = _as_utc(committed_at)+ if moment is None:+ continue+ dated += 1+ if offset is not None:+ with_offset += 1+ if first_at is None or moment < first_at:+ first_at, first_sha = moment, sha+ if last_at is None or moment > last_at:+ last_at = moment+ commit_times.append(moment)+ stamps.append((moment, offset, key))+ if key and (key not in arrival or moment < arrival[key]):+ arrival[key] = moment++ # Author-local only once effectively every row can be shifted; see+ # _LOCAL_TIME_COVERAGE. Mixed clocks would make the matrix a lie.+ local_mode = bool(dated) and (with_offset / dated) >= _LOCAL_TIME_COVERAGE+ timezone_mode = "author_local" if local_mode else "utc"- punch_card = _punch_card_summary(punch, len(commit_times))- velocity = _commit_velocity(commit_times, last_at)+ punch = [[0] * 24 for _ in range(7)]+ commit_days: set[Any] = set()+ day_counts: dict[Any, int] = {}+ per_author_hours: dict[str, list[int]] = {}+ for moment, offset, key in stamps:+ local = moment + timedelta(minutes=offset or 0) if local_mode else moment+ punch[local.weekday()][local.hour] += 1+ day = local.date()+ commit_days.add(day)+ day_counts[day] = day_counts.get(day, 0) + 1+ if local_mode and key and key in human_keys:+ per_author_hours.setdefault(key, [0] * 24)[local.hour] += 1++ busiest_day = None+ if day_counts:+ day, count = max(day_counts.items(), key=lambda kv: kv[1])+ busiest_day = {"date": day.isoformat(), "commits": count}++ months: dict[str, int] = {}+ for moment, offset, _ in stamps:+ local = moment + timedelta(minutes=offset or 0) if local_mode else moment+ month = local.strftime("%Y-%m")+ months[month] = months.get(month, 0) + 1+ busiest_month = None+ if months:+ month, count = max(months.items(), key=lambda kv: kv[1])+ busiest_month = {"month": month, "total": count}++ biggest_commit = next((c for c in top_churn if c["sha"] != first_sha), None)+ widest_commit = next((c for c in top_wide if c["sha"] != first_sha), None)# Prefer the whole-history values stamped on the repo at index time; fall# back to the bounded sample when they're absent (older index, non-git# repo). Age runs from the true first commit to the latest commit we have.- true_total = getattr(repo, "total_commit_count", None)- true_first = getattr(repo, "first_commit_at", None)- true_contributors = getattr(repo, "total_contributor_count", None)- effective_total = true_total if true_total is not None else total+ true_first = _as_utc(getattr(repo, "first_commit_at", None))effective_first = true_first if true_first is not None else first_at- effective_contributors = (- true_contributors if true_contributors is not None else len(contributors)+ effective_total = getattr(repo, "total_commit_count", None)+ effective_contributors = getattr(repo, "total_contributor_count", None)++ arrivals = sorted(+ (+ {"name": display_name.get(k, k), "first_commit_at": _iso(v)}+ for k, v in arrival.items()+ if k in human_keys+ ),+ key=lambda a: a["first_commit_at"] or "",)- age_days = (last_at - effective_first).days if (effective_first and last_at) else Nonereturn {- "total_commits": effective_total,- "agent_commits": agent_total,- "agent_pct": round(agent_total / total * 100.0, 1) if total else 0.0,- "fix_commits": fix_total,- "fix_pct": round(fix_total / total * 100.0, 1) if total else 0.0,- "contributor_count": effective_contributors,- "first_commit_at": _iso(effective_first),- "first_commit_author": getattr(repo, "first_commit_author", None),- "last_commit_at": _iso(last_at),- "age_days": age_days,- "busiest_month": busiest,- "monthly": monthly,- "agent_names": sorted(- ({"name": k, "count": v} for k, v in agent_names.items()),- key=lambda x: -x["count"],- ),+ "origin": {+ "first_commit_at": _iso(effective_first),+ "first_commit_author": getattr(repo, "first_commit_author", None),+ "first_commit_subject": getattr(repo, "first_commit_subject", None),+ "last_commit_at": _iso(last_at),+ "age_days": (+ (last_at - effective_first).days if (effective_first and last_at) else None+ ),+ "total_commits": effective_total if effective_total is not None else total,+ "contributor_count": (+ effective_contributors if effective_contributors is not None else len(contributors)+ ),+ },+ "rhythm": {+ "punch_card": _punch_card_summary(punch, dated, timezone_mode=timezone_mode),+ "velocity": _commit_velocity(commit_times, last_at),+ "busiest_month": busiest_month,+ "busiest_day": busiest_day,+ "longest_streak": _longest_streak(commit_days),+ "active_days": len(commit_days),+ },+ "chronotypes": _chronotypes(per_author_hours, display_name) if local_mode else [],+ "arrivals": arrivals,"biggest_commit": biggest_commit,- "longest_streak": longest_streak,- "punch_card": punch_card,- "velocity": velocity,- "change_risk_mix": risk_mix,+ "widest_commit": widest_commit,}-async def _people(session: AsyncSession, repo_id: str, all_meta: list[Any]) -> dict[str, Any]:- """Ownership concentration: top owners, single-owner files, module silos."""+def _churn_ledger(repo: Any) -> dict[str, Any] | None:+ """Lifetime lines written vs. taken back, from the repo-level totals.++ Deliberately *not* summed from ``git_commits``: that table is bounded to the+ newest N commits, so on a repo with deeper history the sum would be a+ windowed figure presented as a lifetime one. ``None`` when the indexer+ skipped the walk (history too deep) or predates the capture, so the page+ omits the ledger rather than showing a number it can't stand behind.+ """+ added = getattr(repo, "total_lines_added", None)+ deleted = getattr(repo, "total_lines_deleted", None)+ if added is None or deleted is None or added <= 0:+ return None+ return {+ "lines_added": int(added),+ "lines_deleted": int(deleted),+ "net": int(added) - int(deleted),+ "deleted_per_hundred": round(int(deleted) / int(added) * 100.0, 1),+ }+++def _code_half_life(all_meta: list[Any], last_at: str | None) -> int | None:+ """Median days since each file was last touched.++ "Half this codebase hasn't been changed in N days." Anchored to the newest+ commit rather than wall-clock now, so a stale index doesn't inflate it.+ """+ anchor = _as_utc(datetime.fromisoformat(last_at)) if last_at else None+ if anchor is None:+ return None+ ages = sorted(+ (anchor - touched).days+ for m in all_meta+ if (touched := _as_utc(m.last_commit_at)) is not None and touched <= anchor+ )+ if not ages:+ return None+ mid = len(ages) // 2+ return ages[mid] if len(ages) % 2 else (ages[mid - 1] + ages[mid]) // 2+++# ---------------------------------------------------------------------------+# People — repo-level concentration only. Per-person detail is the Contributors+# page's job, and duplicating it here would just be a worse version of it.+# ---------------------------------------------------------------------------+++def _people(all_meta: list[Any]) -> dict[str, Any]:+ """Ownership concentration: owner count, single-owner files, module silos."""owners: dict[str, int] = {}single_owner_files = 0module_owner_files: dict[str, dict[str, int]] = {}@@ -364,12 +579,6 @@ async def _people(session: AsyncSession, repo_id: str, all_meta: list[Any]) -> dbucket = module_owner_files.setdefault(module, {})bucket[m.primary_owner_name] = bucket.get(m.primary_owner_name, 0) + 1- total_files = len(all_meta) or 1- top_owners = sorted(- ({"name": k, "file_count": v, "pct": v / total_files} for k, v in owners.items()),- key=lambda x: -x["file_count"],- )[:8]-silo_count = 0for module, mowners in module_owner_files.items():top = max(mowners.values(), default=0)@@ -392,101 +601,35 @@ async def _people(session: AsyncSession, repo_id: str, all_meta: list[Any]) -> dreturn {"owner_count": len(owners),- "top_owners": top_owners,"single_owner_files": single_owner_files,"silo_count": silo_count,"truck_factor": truck_factor,}-async def _quality(session: AsyncSession, repo_id: str, metrics: list[Any]) -> dict[str, Any]:- """Health KPIs + the defect-validation stat + dead code + doc coverage."""- from repowise.core.analysis.health.defect_accuracy import compute_defect_accuracy- from repowise.core.analysis.health.grading import distribution as health_distribution-- summary = await crud.get_health_summary(session, repo_id)- findings = await crud.get_health_findings(session, repo_id)-- severity = {"critical": 0, "high": 0, "medium": 0, "low": 0}- for f in findings:- s = (f.severity or "").lower()- if s in severity:- severity[s] += 1-- metric_dicts = [- {- "file_path": m.file_path,- "score": m.score,- "nloc": m.nloc,- "has_test_file": m.has_test_file,- "module": m.module,- }- for m in metrics- ]- finding_dicts = [- {- "file_path": f.file_path,- "biomarker_type": f.biomarker_type,- "severity": f.severity,- }- for f in findings- ]- try:- defect_accuracy = compute_defect_accuracy(metric_dicts, finding_dicts)- except Exception:- defect_accuracy = None- try:- dist = health_distribution(metric_dicts)- except Exception:- dist = None-- # Dead code rollup- dead = await crud.get_dead_code_findings(session, repo_id, status="open")- deletable_lines = sum(int(f.lines or 0) for f in dead if f.safe_to_delete)- dead_total = len(dead)-- # Doc coverage (avg page confidence) + page count- avg_conf = float(- await session.scalar(select(func.avg(Page.confidence)).where(Page.repository_id == repo_id))- or 0.0- )- page_count = (- await session.scalar(select(func.count(Page.id)).where(Page.repository_id == repo_id)) or 0- )-- # Test coverage: share of health-metric files that have a test file.- tested = sum(1 for m in metrics if m.has_test_file)- test_coverage_pct = round(tested / len(metrics) * 100.0, 1) if metrics else None-- return {- "average_health": summary.get("average_health"),- "maintainability_average": summary.get("maintainability_average"),- "performance_average": summary.get("performance_average"),- "worst_performer_path": summary.get("worst_performer_path"),- "worst_performer_score": summary.get("worst_performer_score"),- "open_findings": summary.get("open_findings", len(findings)),- "severity_breakdown": severity,- "defect_accuracy": defect_accuracy,- "distribution": dist,- "doc_coverage_pct": avg_conf * 100.0,- "page_count": page_count,- "test_coverage_pct": test_coverage_pct,- "dead_code": {"total_findings": dead_total, "deletable_lines": deletable_lines},- }+# ---------------------------------------------------------------------------+# Records — the superlatives. Several of these are the only headline treatment+# their underlying signal gets anywhere in the app (max CCN is otherwise a+# sortable column; the cycle count has no UI at all).+# ----------------------------------------------------------------------------async def _superlatives(+async def _records(session: AsyncSession, repo_id: str, metrics: list[Any], all_meta: list[Any]) -> dict[str, Any]:- """The fun "biggest / oldest / most" awards, one row each."""+ """The "biggest / oldest / gnarliest / most" awards, one row each."""out: dict[str, Any] = {}- # Largest file by NLOClargest = max(metrics, key=lambda m: m.nloc or 0, default=None)if largest is not None and (largest.nloc or 0) > 0:out["largest_file"] = {"path": largest.file_path, "nloc": int(largest.nloc)}- # Most complex symbol+ # Gnarliest file by cyclomatic complexity. Code Health exposes max_ccn only+ # as a sortable table column, so this is the one place it gets named.+ gnarliest = max(metrics, key=lambda m: m.max_ccn or 0, default=None)+ if gnarliest is not None and (gnarliest.max_ccn or 0) > 0:+ out["gnarliest_file"] = {"path": gnarliest.file_path, "max_ccn": int(gnarliest.max_ccn)}+sym = (await session.execute(select(WikiSymbol.name, WikiSymbol.file_path, WikiSymbol.complexity_estimate)@@ -502,7 +645,6 @@ async def _superlatives("complexity": int(sym[2]),}- # Most-changed file + oldest file (from git metadata)most_changed = max(all_meta, key=lambda m: m.commit_count_total or 0, default=None)if most_changed is not None and (most_changed.commit_count_total or 0) > 0:out["most_changed_file"] = {@@ -511,10 +653,10 @@ async def _superlatives(}dated = [m for m in all_meta if m.first_commit_at is not None]if dated:- oldest = min(dated, key=lambda m: m.first_commit_at)+ oldest = min(dated, key=lambda m: _as_utc(m.first_commit_at))out["oldest_file"] = {"path": oldest.file_path,- "first_commit_at": _iso(oldest.first_commit_at),+ "first_commit_at": _iso(_as_utc(oldest.first_commit_at)),}# Most imported file — highest fan-in among non-test file nodes, the@@ -579,53 +721,12 @@ async def _superlatives(if best_pair and best_pair["count"] > 0:out["strongest_coupling"] = best_pair- return out---async def _dependencies(session: AsyncSession, repo_id: str) -> dict[str, Any] | None:- """Third-party dependency rollup from the manifest-derived rows."""- rows = (- await session.execute(- select(ExternalSystem.ecosystem, ExternalSystem.is_dev_dep, func.count())- .where(ExternalSystem.repository_id == repo_id)- .group_by(ExternalSystem.ecosystem, ExternalSystem.is_dev_dep)- )- ).all()- if not rows:- return None-- total = runtime = dev = 0- ecosystems: dict[str, int] = {}- for eco, is_dev, n in rows:- total += n- if is_dev:- dev += n- else:- runtime += n- key = eco or "other"- ecosystems[key] = ecosystems.get(key, 0) + n-- return {- "total": total,- "runtime": runtime,- "dev": dev,- "ecosystems": sorted(- ({"name": k, "count": v} for k, v in ecosystems.items()),- key=lambda x: -x["count"],- ),- }---async def _graph(session: AsyncSession, repo_id: str) -> dict[str, Any]:- """Dependency-cycle and community structure from the graph snapshot.-- Reads the materialized ``graph_node_membership`` rows (no graph rebuild):- strongly-connected components with ``scc_size > 1`` are import cycles, and- ``symbol_community_id`` groups the natural neighborhoods. Pure aggregate- SQL — a couple of GROUP BY scans, never a row pull."""+ # Largest import cycle — strongly-connected components with more than one+ # member. One aggregate over the materialized membership rows (no graph+ # rebuild). Nothing else in the app surfaces cycles at all.scc_rows = (await session.execute(- select(GraphNodeMembership.scc_id, func.count())+ select(func.count()).where(GraphNodeMembership.repository_id == repo_id,GraphNodeMembership.scc_size > 1,@@ -633,54 +734,41 @@ async def _graph(session: AsyncSession, repo_id: str) -> dict[str, Any]:.group_by(GraphNodeMembership.scc_id))).all()- cycle_clusters = len(scc_rows)- files_in_cycles = sum(int(n) for _, n in scc_rows)- largest_cycle = max((int(n) for _, n in scc_rows), default=0)-- community_count = (- await session.scalar(- select(func.count(func.distinct(GraphNodeMembership.symbol_community_id))).where(- GraphNodeMembership.repository_id == repo_id,- GraphNodeMembership.symbol_community_id.is_not(None),- )- )- or 0- )-- return {- "cycle_clusters": cycle_clusters,- "files_in_cycles": files_in_cycles,- "largest_cycle": largest_cycle,- "community_count": int(community_count),- }---async def _build(session: AsyncSession, repo_id: str) -> dict[str, Any]:- """The knowledge base's own build stats — the wiki bragging about itself.+ if scc_rows:+ out["largest_cycle"] = {+ "files": max(int(n) for (n,) in scc_rows),+ "cycle_count": len(scc_rows),+ }- Pages come from the ``wiki_pages`` count; tokens and cost from the- ``llm_costs`` ledger written during generation. Single aggregate scan;- every field degrades to 0 when generation was index-only (no LLM spend)."""- page_count = (- await session.scalar(select(func.count(Page.id)).where(Page.repository_id == repo_id)) or 0- )- cost_row = (+ # Symbol-shape shares: what fraction of this codebase is async, and what+ # fraction carries a docstring. One aggregate row, three counters.+ shape = (await session.execute(select(- func.coalesce(func.sum(LlmCost.input_tokens + LlmCost.output_tokens), 0),- func.coalesce(func.sum(LlmCost.cost_usd), 0.0),- func.count(LlmCost.id),- ).where(LlmCost.repository_id == repo_id)+ func.count(WikiSymbol.id),+ func.sum(func.cast(WikiSymbol.is_async, Integer)),+ func.count(WikiSymbol.docstring),+ ).where(WikiSymbol.repository_id == repo_id))- ).one()- total_tokens, cost_usd, op_count = cost_row+ ).one_or_none()+ if shape and shape[0]:+ total_symbols = int(shape[0])+ async_count = int(shape[1] or 0)+ documented = int(shape[2] or 0)+ out["symbol_shape"] = {+ "total": total_symbols,+ "async_count": async_count,+ "async_pct": round(async_count / total_symbols * 100.0, 1),+ "documented_count": documented,+ "documented_pct": round(documented / total_symbols * 100.0, 1),+ }- return {- "page_count": int(page_count),- "total_tokens": int(total_tokens or 0),- "cost_usd": round(float(cost_usd or 0.0), 2),- "llm_operations": int(op_count or 0),- }+ return out+++# ---------------------------------------------------------------------------+# Endpoint+# ---------------------------------------------------------------------------@router.get("/{repo_id}/stats/highlights")@@ -688,60 +776,68 @@ async def stats_highlights(repo_id: str,session: AsyncSession = Depends(get_db_session), # noqa: B008) -> dict:- """Everything the Stats ("By the Numbers") page needs, in one call."""+ """Everything the Stats ("By the Numbers") page needs, in one call.++ Three narrow column selects (health metrics, git metadata, commits) plus a+ handful of aggregates. Deliberately no health summary, findings, dead-code,+ dependency, decision or cost reads: every one of those belongs to a page+ that already owns the subject, and pulling them here cost the page its two+ slowest queries for numbers shown better elsewhere.+ """repo = await crud.get_repository(session, repo_id)if repo is None:raise HTTPException(status_code=404, detail="Repository not found")- metrics = await crud.get_health_metrics(session, repo_id)- all_meta = list(- (await session.execute(select(GitMetadata).where(GitMetadata.repository_id == repo_id)))- .scalars()- .all()- )-- decision_count = (- await session.scalar(- select(func.count(DecisionRecord.id)).where(DecisionRecord.repository_id == repo_id)+ # Narrow selects, not whole ORM rows: the page needs four health columns and+ # seven git-metadata ones out of the ~20 each table carries, and the unread+ # remainder includes the large JSON blobs.+ metrics = (+ await session.execute(+ select(+ HealthFileMetric.file_path,+ HealthFileMetric.nloc,+ HealthFileMetric.module,+ HealthFileMetric.max_ccn,+ ).where(HealthFileMetric.repository_id == repo_id))- or 0- )- active_decisions = (- await session.scalar(- select(func.count(DecisionRecord.id)).where(- DecisionRecord.repository_id == repo_id,- DecisionRecord.status == "active",- )+ ).all()+ all_meta = (+ await session.execute(+ select(+ GitMetadata.file_path,+ GitMetadata.primary_owner_name,+ GitMetadata.bus_factor,+ GitMetadata.commit_count_total,+ GitMetadata.first_commit_at,+ GitMetadata.last_commit_at,+ GitMetadata.co_change_partners_json,+ ).where(GitMetadata.repository_id == repo_id))- or 0- )+ ).all()- activity = await _activity(session, repo_id, repo)- superlatives = await _superlatives(session, repo_id, metrics, all_meta)- # Computed on _activity's commit scan for performance, but they are awards- # so they belong under superlatives in the payload.- for key in ("biggest_commit", "longest_streak"):- value = activity.pop(key, None)+ commits = await _commit_pass(session, repo_id, repo)+ records = await _records(session, repo_id, metrics, all_meta)+ # Computed on the commit scan for cost, but they are awards, so the payload+ # files them under records.+ for key in ("biggest_commit", "widest_commit"):+ value = commits.pop(key, None)if value:- superlatives[key] = value+ records[key] = value++ rhythm = commits["rhythm"]+ rhythm["code_half_life_days"] = _code_half_life(all_meta, commits["origin"]["last_commit_at"])++ people = _people(all_meta)+ people["contributor_count"] = commits["origin"]["contributor_count"]+ people["chronotypes"] = commits.pop("chronotypes", [])+ people["arrivals"] = commits.pop("arrivals", [])return {- "repo": {- "id": repo.id,- "name": repo.name,- "default_branch": repo.default_branch,- "head_commit": repo.head_commit,- },+ "repo": {"id": repo.id, "name": repo.name},"scale": await _scale(session, repo_id, metrics),- "activity": activity,- "people": await _people(session, repo_id, all_meta),- "quality": await _quality(session, repo_id, metrics),- "knowledge": {- "decision_count": decision_count,- "active_decision_count": active_decisions,- },- "dependencies": await _dependencies(session, repo_id),- "graph": await _graph(session, repo_id),- "build": await _build(session, repo_id),- "superlatives": superlatives,+ "origin": commits["origin"],+ "churn": _churn_ledger(repo),+ "rhythm": rhythm,+ "people": people,+ "records": records,}diff --git a/packages/types/src/overview.ts b/packages/types/src/overview.tsindex 1c7c37e97..9f32d8b6f 100644--- a/packages/types/src/overview.ts+++ b/packages/types/src/overview.ts@@ -5,6 +5,7 @@*/import type { Hotspot } from "./git.js";+import type { DefectAccuracy } from "./health.js";export interface OverviewRepoMeta {id: string;@@ -56,6 +57,10 @@ export interface OverviewHealthHistoryPoint {export interface OverviewHealth {average_health: number | null;+ /** Backtested precision of the defect ranking — the "can you trust this+ * score?" panel on the health card. Optional so an older server payload+ * simply omits the panel. */+ defect_accuracy?: DefectAccuracy | null;hotspot_health: number | null;worst_performer_path: string | null;worst_performer_score: number | null;diff --git a/packages/types/src/stats.ts b/packages/types/src/stats.tsindex b9e3992a5..eb8e25f0f 100644--- a/packages/types/src/stats.ts+++ b/packages/types/src/stats.ts@@ -2,9 +2,14 @@* Data contract for the repo Stats ("By the Numbers") page.** Mirrors the payload from `GET /api/repos/{repo_id}/stats/highlights`- * (packages/server/.../routers/stats.py). Every section is independently- * built server-side and degrades to null/empty rather than failing the page,- * so most leaf fields are nullable.+ * (packages/server/.../routers/stats.py). Scope is defined by subtraction:+ * only signals no other page in the app already owns. Health scores belong to+ * Code Health, commit volume and categories to Commits, per-person ownership to+ * Contributors, dependencies and communities to Architecture — none of those+ * appear here.+ *+ * Every section is independently built server-side and degrades to null/empty+ * rather than failing the page, so most leaf fields are nullable.*/export interface StatsSizeClass {@@ -21,7 +26,6 @@ export interface StatsLanguage {export interface StatsScale {file_count: number;symbol_count: number;- entry_point_count: number;module_count: number;total_nloc: number;language_count: number;@@ -29,19 +33,43 @@ export interface StatsScale {size_class: StatsSizeClass;}-export interface StatsMonthlyBucket {- month: string;- total: number;- agent: number;+/** The repo's founding moment — its opening line, not just a date. */+export interface StatsOrigin {+ first_commit_at: string | null;+ /** Founding author (root commit). Null for older indexes / non-git repos. */+ first_commit_author: string | null;+ /** The root commit's subject. Null until an index captures it. */+ first_commit_subject: string | null;+ last_commit_at: string | null;+ age_days: number | null;+ total_commits: number;+ contributor_count: number;}-export interface StatsAgentName {- name: string;- count: number;+/**+ * Lifetime lines written vs. taken back.+ *+ * Read from repo-level totals captured at index time, never summed from the+ * commit table — that table is bounded to the newest N commits, so summing it+ * would present a windowed figure as a lifetime one. Null when the history was+ * too deep to walk or the index predates the capture.+ */+export interface StatsChurn {+ lines_added: number;+ lines_deleted: number;+ net: number;+ /** Lines deleted per hundred written. The shareable ratio. */+ deleted_per_hundred: number;}-/** Coding-rhythm heatmap: commit counts by weekday (0=Monday) x hour (0-23,- * in the stored UTC), with the human-readable hooks the hero renders. */+/**+ * Coding-rhythm heatmap: commit counts by weekday (0=Monday) x hour (0-23).+ *+ * `timezone_mode` says which clock the matrix is drawn in. `author_local` means+ * every commit was shifted by its author's own UTC offset — the honest version+ * of "when do people work". `utc` is the fallback for indexes written before+ * the offset was captured; it resolves itself on the next `repowise update`.+ */export interface StatsPunchCard {/** 7 rows (Mon..Sun) x 24 columns (hours). */matrix: number[][];@@ -51,6 +79,7 @@ export interface StatsPunchCard {busiest_weekday: number | null;peak_hour: number | null;total: number;+ timezone_mode: "author_local" | "utc";}/** Commit momentum: the 90 days ending at the newest commit vs the 90 before. */@@ -61,111 +90,63 @@ export interface StatsVelocity {pct_change: number | null;}-/** Calibrated just-in-time change-risk tally across the sampled commits. */-export interface StatsChangeRiskMix {- low: number;- moderate: number;- high: number;-}--export interface StatsActivity {- total_commits: number;- agent_commits: number;- agent_pct: number;- fix_commits: number;- fix_pct: number;- contributor_count: number;- first_commit_at: string | null;- /** Founding author (root commit). Null for older indexes / non-git repos. */- first_commit_author: string | null;- last_commit_at: string | null;- age_days: number | null;- busiest_month: StatsMonthlyBucket | null;- monthly: StatsMonthlyBucket[];- agent_names: StatsAgentName[];+/** The time-shape of the work. Nothing else in the app has a clock. */+export interface StatsRhythm {punch_card: StatsPunchCard;velocity: StatsVelocity;- change_risk_mix: StatsChangeRiskMix;+ busiest_month: { month: string; total: number } | null;+ busiest_day: { date: string; commits: number } | null;+ /** Longest run of consecutive days with at least one commit. */+ longest_streak: { days: number; start: string; end: string } | null;+ /** Distinct calendar days carrying at least one commit. */+ active_days: number;+ /** Median days since each file was last touched, anchored to the newest+ * commit rather than now, so a stale index doesn't inflate it. */+ code_half_life_days: number | null;+}++/** A contributor's commit-hour habit. Only emitted in author-local mode —+ * awarding "night owl" off UTC would just name whoever lives furthest east. */+export interface StatsChronotype {+ name: string;+ commits: number;+ peak_hour: number;+ label: "night_owl" | "early_bird" | "daylight";+ night_pct: number;+ early_pct: number;}-export interface StatsOwner {+export interface StatsArrival {name: string;- file_count: number;- pct: number;+ first_commit_at: string | null;}+/**+ * Repo-level concentration only. Per-person detail (ownership share, hotspots+ * owned, dead-code burden) is the Contributors page's job.+ */export interface StatsPeople {owner_count: number;- top_owners: StatsOwner[];+ contributor_count: number;single_owner_files: number;silo_count: number;/** Fewest primary owners who together hold >50% of owned files. 1 means a* single person owns most of the codebase. Null when no ownership data. */truck_factor: number | null;+ chronotypes: StatsChronotype[];+ /** Contributors ordered by their first commit — the arrivals timeline. */+ arrivals: StatsArrival[];}-export interface StatsSeverityBreakdown {- critical: number;- high: number;- medium: number;- low: number;-}--export interface StatsDefectAccuracy {- k: number;- hits: number;- precision: number;- base_rate: number;- lift: number | null;- window_days: number;- scored_files: number;- defect_files: number;-}--export interface StatsDistributionBand {- files: number;- nloc: number;- pct: number;-}--export interface StatsDistribution {- total_files: number;- total_nloc: number;- bands: {- healthy: StatsDistributionBand;- warning: StatsDistributionBand;- alert: StatsDistributionBand;- };-}--export interface StatsDeadCode {- total_findings: number;- deletable_lines: number;-}--export interface StatsQuality {- average_health: number | null;- maintainability_average: number | null;- performance_average: number | null;- worst_performer_path: string | null;- worst_performer_score: number | null;- open_findings: number;- severity_breakdown: StatsSeverityBreakdown;- defect_accuracy: StatsDefectAccuracy | null;- distribution: StatsDistribution | null;- doc_coverage_pct: number;- page_count: number;- test_coverage_pct: number | null;- dead_code: StatsDeadCode;-}--export interface StatsKnowledge {- decision_count: number;- active_decision_count: number;-}--export interface StatsSuperlatives {+/**+ * The superlatives. Several are the only headline treatment their underlying+ * signal gets anywhere: max CCN is otherwise a sortable column, and import+ * cycles have no other UI at all.+ */+export interface StatsRecords {largest_file?: { path: string; nloc: number };+ /** Highest cyclomatic complexity in any single file. */+ gnarliest_file?: { path: string; max_ccn: number };most_complex_symbol?: { name: string; file_path: string; complexity: number };most_changed_file?: { path: string; commit_count: number };oldest_file?: { path: string; first_commit_at: string | null };@@ -173,6 +154,16 @@ export interface StatsSuperlatives {* is then "most imported"; without it, it degrades to the PageRank pick. */most_central_file?: { path: string; pagerank: number; import_count?: number };strongest_coupling?: { a: string; b: string; count: number };+ /** Biggest strongly-connected import cycle, and how many exist. */+ largest_cycle?: { files: number; cycle_count: number };+ /** What fraction of this codebase is async, and what fraction is documented. */+ symbol_shape?: {+ total: number;+ async_count: number;+ async_pct: number;+ documented_count: number;+ documented_pct: number;+ };/** Largest non-initial commit by churn (added + deleted lines). */biggest_commit?: {sha: string;@@ -180,59 +171,27 @@ export interface StatsSuperlatives {lines_changed: number;files_changed: number;};- /** Longest run of consecutive days with at least one commit (UTC dates). */- longest_streak?: { days: number; start: string; end: string };-}--export interface StatsEcosystem {- name: string;- count: number;-}--export interface StatsDependencies {- total: number;- runtime: number;- dev: number;- ecosystems: StatsEcosystem[];-}--/** Import-graph structure: dependency cycles and natural communities, read- * from the materialized graph snapshot (no rebuild). */-export interface StatsGraph {- /** Strongly-connected components with >1 member — circular import clusters. */- cycle_clusters: number;- files_in_cycles: number;- largest_cycle: number;- community_count: number;-}--/** The knowledge base's own build cost — the wiki bragging about itself. */-export interface StatsBuild {- page_count: number;- total_tokens: number;- cost_usd: number;- llm_operations: number;+ /** Non-initial commit touching the most files. */+ widest_commit?: {+ sha: string;+ subject: string;+ files_changed: number;+ lines_changed: number;+ };}export interface StatsRepo {id: string;name: string;- default_branch: string;- head_commit: string | null;}export interface StatsHighlights {repo: StatsRepo;scale: StatsScale;- activity: StatsActivity;+ origin: StatsOrigin;+ /** Null when lifetime churn could not be captured. */+ churn: StatsChurn | null;+ rhythm: StatsRhythm;people: StatsPeople;- quality: StatsQuality;- knowledge: StatsKnowledge;- /** Optional so older payloads (hosted exporters) degrade gracefully. */- dependencies?: StatsDependencies | null;- /** Optional so older payloads (hosted exporters) degrade gracefully. */- graph?: StatsGraph | null;- /** Optional so older payloads (hosted exporters) degrade gracefully. */- build?: StatsBuild | null;- superlatives: StatsSuperlatives;+ records: StatsRecords;}diff --git a/packages/ui/__tests__/stats/punch-card.test.tsx b/packages/ui/__tests__/stats/punch-card.test.tsxindex 344bc24e3..cb38d9f7b 100644--- a/packages/ui/__tests__/stats/punch-card.test.tsx+++ b/packages/ui/__tests__/stats/punch-card.test.tsx@@ -16,7 +16,8 @@ function makePunchCard(): StatsPunchCard {peak: { weekday: 2, hour: 14, count: 6 },busiest_weekday: 2,peak_hour: 14,- total: 10,+ timezone_mode: "author_local" as const,+ total: 10,};}diff --git a/packages/ui/__tests__/stats/size-class-hero.test.tsx b/packages/ui/__tests__/stats/size-class-hero.test.tsxindex 167beab6d..6a52761b9 100644--- a/packages/ui/__tests__/stats/size-class-hero.test.tsx+++ b/packages/ui/__tests__/stats/size-class-hero.test.tsx@@ -7,7 +7,6 @@ function makeScale(overrides: Partial<StatsScale> = {}): StatsScale {return {file_count: 1_234_567,symbol_count: 98_432,- entry_point_count: 12,module_count: 80,total_nloc: 4_200_000,language_count: 3,@@ -22,30 +21,29 @@ function makeScale(overrides: Partial<StatsScale> = {}): StatsScale {}describe("SizeClassHero", () => {- it("renders large file and symbol counts compactly", () => {- render(<SizeClassHero scale={makeScale()} />);+ it("names the size class and its blurb", () => {+ render(<SizeClassHero scale={makeScale()} repoName="acme" />);- expect(screen.getByText("1.2M")).toBeInTheDocument();- expect(screen.getByText("98.4K")).toBeInTheDocument();- expect(screen.queryByText("1,234,567")).not.toBeInTheDocument();- expect(screen.queryByText("98,432")).not.toBeInTheDocument();+ expect(screen.getByRole("heading", { name: "Metropolis" })).toBeInTheDocument();+ expect(screen.getByText("A large codebase.")).toBeInTheDocument();+ expect(screen.getByText("acme")).toBeInTheDocument();});- it("renders small counts verbatim", () => {- render(<SizeClassHero scale={makeScale({ file_count: 999, symbol_count: 42 })} />);+ it("renders no figures of its own", () => {+ // The headline counts moved to StatRibbon. Rendering them here too put a+ // boxed number directly above the hairline row carrying the same number,+ // so the hero is now identity-only and this guards against them creeping+ // back in.+ render(<SizeClassHero scale={makeScale()} />);- expect(screen.getByText("999")).toBeInTheDocument();- expect(screen.getByText("42")).toBeInTheDocument();- expect(screen.getByText("3")).toBeInTheDocument();+ expect(screen.queryByText("1.2M")).not.toBeInTheDocument();+ expect(screen.queryByText("98.4K")).not.toBeInTheDocument();+ expect(screen.queryByText(/Lines of code/i)).not.toBeInTheDocument();+ expect(screen.queryByText(/Symbols/i)).not.toBeInTheDocument();});- it("never clips a figure to an ellipsis", () => {- render(<SizeClassHero scale={makeScale({ total_nloc: 485_300 })} />);-- const value = screen.getByText("485.3K");- expect(value).toHaveClass("whitespace-nowrap");- expect(value).not.toHaveClass("truncate");- // The tile must not be allowed to shrink below its figure.- expect(value.parentElement).not.toHaveClass("min-w-0");+ it("omits the eyebrow when no repo name is given", () => {+ const { container } = render(<SizeClassHero scale={makeScale()} />);+ expect(container.querySelector(".tracking-\\[0\\.2em\\]")).toBeNull();});});diff --git a/packages/ui/__tests__/stats/stat-ribbon.test.tsx b/packages/ui/__tests__/stats/stat-ribbon.test.tsxnew file mode 100644index 000000000..3cf20e09e--- /dev/null+++ b/packages/ui/__tests__/stats/stat-ribbon.test.tsx@@ -0,0 +1,45 @@+import { render, screen } from "@testing-library/react";+import { describe, expect, it } from "vitest";+import { StatRibbon } from "../../src/stats/stat-ribbon.js";++describe("StatRibbon", () => {+ it("renders each stat as a labelled definition pair", () => {+ render(+ <StatRibbon+ stats={[+ { label: "Lines of code", value: "419.6K" },+ { label: "Files", value: "3,146" },+ ]}+ />,+ );++ expect(screen.getByText("Lines of code")).toBeInTheDocument();+ expect(screen.getByText("419.6K")).toBeInTheDocument();+ expect(screen.getByText("Files")).toBeInTheDocument();+ expect(screen.getByText("3,146")).toBeInTheDocument();+ });++ it("lines figures up as tabular numerals", () => {+ render(<StatRibbon stats={[{ label: "Files", value: "3,146" }]} />);+ expect(screen.getByText("3,146")).toHaveClass("tabular-nums");+ });++ it("drops stats with no value rather than rendering an empty cell", () => {+ render(+ <StatRibbon+ stats={[+ { label: "Files", value: "3,146" },+ { label: "Modules", value: "" },+ ]}+ />,+ );++ expect(screen.getByText("Files")).toBeInTheDocument();+ expect(screen.queryByText("Modules")).not.toBeInTheDocument();+ });++ it("renders nothing at all when every stat is empty", () => {+ const { container } = render(<StatRibbon stats={[{ label: "Files", value: "" }]} />);+ expect(container).toBeEmptyDOMElement();+ });+});diff --git a/packages/ui/src/stats/activity-trend-chart.tsx b/packages/ui/src/stats/activity-trend-chart.tsxdeleted file mode 100644index c077e4269..000000000--- a/packages/ui/src/stats/activity-trend-chart.tsx+++ /dev/null@@ -1,75 +0,0 @@-"use client";--import { BarChart, Bar, XAxis, YAxis, Tooltip, ResponsiveContainer } from "recharts";-import type { StatsMonthlyBucket } from "@repowise-dev/types/stats";-import { Card, CardContent, CardHeader, CardTitle } from "../ui/card";--interface ActivityTrendChartProps {- monthly: StatsMonthlyBucket[];- title?: string;-}--/** Stacked monthly commit volume, split into agent-authored vs human. The- * "when did AI start writing this code" view. */-export function ActivityTrendChart({ monthly, title = "Commits per month" }: ActivityTrendChartProps) {- if (monthly.length === 0) return null;-- const data = monthly.map((b) => ({- month: b.month,- human: Math.max(b.total - b.agent, 0),- agent: b.agent,- }));-- return (- <Card className="h-full">- <CardHeader className="pb-2">- <CardTitle className="text-sm">{title}</CardTitle>- </CardHeader>- <CardContent className="pt-0">- <div className="h-[220px] w-full">- <ResponsiveContainer width="100%" height="100%">- <BarChart data={data} margin={{ top: 8, right: 8, bottom: 0, left: -16 }}>- <XAxis- dataKey="month"- tick={{ fontSize: 10, fill: "var(--color-text-tertiary)" }}- tickLine={false}- axisLine={false}- interval="preserveStartEnd"- minTickGap={24}- />- <YAxis- tick={{ fontSize: 10, fill: "var(--color-text-tertiary)" }}- tickLine={false}- axisLine={false}- allowDecimals={false}- width={40}- />- <Tooltip- cursor={{ fill: "var(--color-bg-muted)", opacity: 0.4 }}- contentStyle={{- background: "var(--color-bg-overlay)",- border: "1px solid var(--color-border-default)",- borderRadius: 8,- fontSize: 11,- color: "var(--color-text-primary)",- }}- />- <Bar dataKey="human" stackId="c" fill="var(--color-accent-primary)" radius={[0, 0, 0, 0]} />- <Bar dataKey="agent" stackId="c" fill="var(--color-info)" radius={[2, 2, 0, 0]} />- </BarChart>- </ResponsiveContainer>- </div>- <div className="mt-2 flex items-center gap-4 text-[11px] text-[var(--color-text-secondary)]">- <span className="flex items-center gap-1.5">- <span className="h-2 w-2 rounded-full" style={{ background: "var(--color-accent-primary)" }} />- Human- </span>- <span className="flex items-center gap-1.5">- <span className="h-2 w-2 rounded-full" style={{ background: "var(--color-info)" }} />- Agent- </span>- </div>- </CardContent>- </Card>- );-}diff --git a/packages/ui/src/stats/arrivals-timeline.tsx b/packages/ui/src/stats/arrivals-timeline.tsxnew file mode 100644index 000000000..a41ee8617--- /dev/null+++ b/packages/ui/src/stats/arrivals-timeline.tsx@@ -0,0 +1,62 @@+"use client";++import * as React from "react";+import type { StatsArrival } from "@repowise-dev/types/stats";+import { formatDate, formatNumber } from "../lib/format";++const INITIAL_SHOWN = 12;++/**+ * Contributors in the order they first showed up.+ *+ * The Contributors page shows one person's join date on their own profile, so+ * the sequence — who was here first, when the team grew — exists nowhere else.+ * Reads as a roll rather than a chart: the dates are the interesting part, and a+ * timeline axis would compress the early years into nothing on a repo whose+ * hiring accelerated.+ */+export function ArrivalsTimeline({ arrivals }: { arrivals: StatsArrival[] }) {+ const [expanded, setExpanded] = React.useState(false);+ const dated = (arrivals ?? []).filter((a) => a.first_commit_at);+ if (dated.length === 0) return null;++ const shown = expanded ? dated : dated.slice(0, INITIAL_SHOWN);+ const hidden = dated.length - shown.length;++ return (+ <section aria-label="Contributor arrivals" className="flex flex-col gap-4">+ <div className="flex flex-wrap items-baseline justify-between gap-x-4 gap-y-1">+ <h3 className="text-base font-semibold text-[var(--color-text-primary)]">Arrivals</h3>+ <span className="font-mono text-[11px] tabular-nums text-[var(--color-text-tertiary)]">+ {formatNumber(dated.length)} contributors+ </span>+ </div>++ <ol className="flex flex-col divide-y divide-[var(--color-border-default)] border-y border-[var(--color-border-default)]">+ {shown.map((a, i) => (+ <li key={`${a.name}-${a.first_commit_at}`} className="flex items-baseline gap-3 py-2.5">+ <span className="w-6 shrink-0 font-mono text-[11px] tabular-nums text-[var(--color-text-tertiary)]">+ {i + 1}+ </span>+ <span className="min-w-0 flex-1 truncate text-sm text-[var(--color-text-primary)]">+ {a.name}+ </span>+ <span className="shrink-0 font-mono text-[11px] tabular-nums text-[var(--color-text-secondary)]">+ {formatDate(a.first_commit_at as string)}+ </span>+ </li>+ ))}+ </ol>++ {hidden > 0 && (+ <button+ type="button"+ onClick={() => setExpanded(true)}+ className="self-start rounded-md px-1 text-xs font-medium text-[var(--color-accent-primary)] underline-offset-4 hover:underline focus-visible:outline focus-visible:outline-2 focus-visible:outline-offset-2 focus-visible:outline-[var(--color-accent-primary)]"+ >+ Show {formatNumber(hidden)} more+ </button>+ )}+ </section>+ );+}diff --git a/packages/ui/src/stats/chronotype-list.tsx b/packages/ui/src/stats/chronotype-list.tsxnew file mode 100644index 000000000..07c6bd334--- /dev/null+++ b/packages/ui/src/stats/chronotype-list.tsx@@ -0,0 +1,74 @@+import { Moon, Sun, Sunrise } from "lucide-react";+import type { StatsChronotype } from "@repowise-dev/types/stats";+import { formatNumber } from "../lib/format";++const LABELS: Record<+ StatsChronotype["label"],+ { text: string; icon: typeof Moon; color: string }+> = {+ night_owl: { text: "Night owl", icon: Moon, color: "var(--color-info)" },+ early_bird: { text: "Early bird", icon: Sunrise, color: "var(--color-warning)" },+ daylight: { text: "Daylight", icon: Sun, color: "var(--color-text-tertiary)" },+};++function hourLabel(h: number): string {+ const period = h < 12 ? "am" : "pm";+ const twelve = h % 12 === 0 ? 12 : h % 12;+ return `${twelve}${period}`;+}++/**+ * When each frequent contributor actually commits.+ *+ * Only rendered when the index carries per-commit UTC offsets — in UTC mode the+ * "night owl" award would just go to whoever lives furthest east, so the server+ * withholds the data entirely rather than let the UI publish a timezone+ * artifact as a personality trait.+ *+ * The bar is the person's own 24-hour histogram, normalised to their own peak,+ * so someone with 600 commits and someone with 20 are compared on shape rather+ * than volume.+ */+export function ChronotypeList({ people }: { people: StatsChronotype[] }) {+ if (!people || people.length === 0) return null;++ return (+ <section aria-label="Commit-hour habits" className="flex flex-col gap-4">+ <div className="flex flex-wrap items-baseline justify-between gap-x-4 gap-y-1">+ <h3 className="text-base font-semibold text-[var(--color-text-primary)]">+ Night owls & early birds+ </h3>+ <span className="font-mono text-[11px] uppercase tracking-[0.1em] text-[var(--color-text-tertiary)]">+ Each person's local time+ </span>+ </div>++ <ul className="flex flex-col divide-y divide-[var(--color-border-default)] border-y border-[var(--color-border-default)]">+ {people.map((p) => {+ const meta = LABELS[p.label] ?? LABELS.daylight;+ const Icon = meta.icon;+ return (+ <li key={p.name} className="flex items-center gap-4 py-3">+ <div className="min-w-0 flex-1">+ <p className="truncate text-sm font-medium text-[var(--color-text-primary)]">+ {p.name}+ </p>+ <p className="text-xs text-[var(--color-text-tertiary)] tabular-nums">+ {formatNumber(p.commits)} commits · peaks around {hourLabel(p.peak_hour)}+ </p>+ </div>+ <span+ className="flex shrink-0 items-center gap-1.5 rounded-full border px-2.5 py-1 text-[11px] font-medium"+ style={{ color: meta.color, borderColor: meta.color }}+ title={`${p.night_pct}% of commits between 10pm and 5am`}+ >+ <Icon className="h-3.5 w-3.5" />+ {meta.text}+ </span>+ </li>+ );+ })}+ </ul>+ </section>+ );+}diff --git a/packages/ui/src/stats/churn-ledger.tsx b/packages/ui/src/stats/churn-ledger.tsxnew file mode 100644index 000000000..277a8bb36--- /dev/null+++ b/packages/ui/src/stats/churn-ledger.tsx@@ -0,0 +1,90 @@+import type { StatsChurn } from "@repowise-dev/types/stats";+import { formatNumber } from "../lib/format";++/**+ * Lifetime lines written, and how many were later taken back.+ *+ * One bar, three readings. The full width is everything ever written; the+ * hatched tail is what was deleted; what remains is what survived. Deletion+ * reads as erosion of the whole rather than as a second competing bar, which is+ * what a diverging two-sided chart would imply — and "taken back" is a share of+ * the work, not its opposite.+ *+ * The ratio line underneath is the point of the whole component.+ */+export function ChurnLedger({ data }: { data: StatsChurn }) {+ if (!data || data.lines_added <= 0) return null;++ // Guard the pathological case where a rewrite deletes more than the running+ // total ever added: clamp so the bar can't invert or overflow its track.+ const deleted = Math.max(0, Math.min(data.lines_deleted, data.lines_added));+ const standing = data.lines_added - deleted;+ const standingPct = (standing / data.lines_added) * 100;++ const figures = [+ { label: "Written", value: data.lines_added, tone: "var(--color-text-primary)" },+ { label: "Taken back", value: data.lines_deleted, tone: "var(--color-error)" },+ // `standing`, not the server's `net`: a repo that has deleted more than it+ // ever added (a big vendored-tree removal) drives net negative, which would+ // read as "-40,000 still standing" beside a bar showing zero.+ { label: "Still standing", value: standing, tone: "var(--color-success)" },+ ];++ return (+ <section aria-label="Lifetime line churn" className="flex flex-col gap-3.5">+ <div+ className="flex h-11 overflow-hidden rounded-md border border-[var(--color-border-default)]"+ role="img"+ aria-label={`${formatNumber(data.lines_added)} lines written, of which ${formatNumber(+ data.lines_deleted,+ )} were later deleted`}+ >+ <div+ className="relative shrink-0 bg-[var(--color-success-muted,rgba(47,125,82,0.12))]"+ style={{ width: `${standingPct}%` }}+ >+ <span className="absolute inset-0 flex items-center truncate px-3 font-mono text-[11px] uppercase tracking-[0.06em] text-[var(--color-success)]">+ {formatNumber(standing)} standing+ </span>+ </div>+ <div+ className="relative flex-1 border-l border-[var(--color-border-default)]"+ style={{+ // Hatching rather than a flat fill: the deleted share is a texture+ // over the same track, not a separate quantity beside it.+ background:+ "repeating-linear-gradient(-45deg, var(--color-error-muted, rgba(163,58,50,0.14)) 0 6px, transparent 6px 12px)",+ }}+ >+ <span className="absolute inset-0 flex items-center truncate px-3 font-mono text-[11px] uppercase tracking-[0.06em] text-[var(--color-error)]">+ {formatNumber(deleted)} cut+ </span>+ </div>+ </div>++ <div className="flex flex-wrap gap-x-10 gap-y-3">+ {figures.map((f) => (+ <div key={f.label}>+ <p className="font-mono text-[10px] uppercase tracking-[0.12em] text-[var(--color-text-tertiary)]">+ {f.label}+ </p>+ <p+ className="text-2xl font-semibold tabular-nums leading-tight"+ style={{ color: f.tone }}+ >+ {formatNumber(f.value)}+ </p>+ </div>+ ))}+ </div>++ <p className="text-[15px] text-[var(--color-text-primary)]">+ For every 100 lines ever written here,{" "}+ <strong className="font-semibold text-[var(--color-accent-primary)]">+ {Math.round(data.deleted_per_hundred)} were taken back+ </strong>+ .+ </p>+ </section>+ );+}diff --git a/packages/ui/src/stats/index.ts b/packages/ui/src/stats/index.tsindex b006d4400..6d4616768 100644--- a/packages/ui/src/stats/index.ts+++ b/packages/ui/src/stats/index.ts@@ -6,8 +6,12 @@ export {type StatCalloutProps,type CalloutTone,} from "./stat-callout";-export { SuperlativesGrid } from "./superlatives-grid";-export { ActivityTrendChart } from "./activity-trend-chart";+export { OriginBlock } from "./origin-block";+export { ChurnLedger } from "./churn-ledger";+export { StatRibbon, type RibbonStat } from "./stat-ribbon";+export { RecordsGrid } from "./records-grid";+export { ChronotypeList } from "./chronotype-list";+export { ArrivalsTimeline } from "./arrivals-timeline";export { PunchCard } from "./punch-card";export {WEEKEND_PRESETS,diff --git a/packages/ui/src/stats/origin-block.tsx b/packages/ui/src/stats/origin-block.tsxnew file mode 100644index 000000000..b27f51bba--- /dev/null+++ b/packages/ui/src/stats/origin-block.tsx@@ -0,0 +1,77 @@+import type { StatsOrigin } from "@repowise-dev/types/stats";+import { formatAgeDays, formatNumber } from "../lib/format";++/** Absolute date in UTC.+ *+ * Deliberately not the browser's locale timezone: this is the repo's founding+ * moment, a fact about the project rather than about the reader, and rendering+ * it locally would show two different birthdays to two people looking at the+ * same repo (and slip a day for anyone far enough east or west). */+function foundingDate(iso: string): string {+ return new Date(iso).toLocaleDateString(undefined, {+ day: "numeric",+ month: "short",+ year: "numeric",+ timeZone: "UTC",+ });+}++/**+ * The repo's opening line, set as a commit rather than a statistic.+ *+ * Every other date on this page is a measurement; this one is the project's+ * origin story, so it gets the page's only quotation treatment — an amber rule,+ * the subject in mono at display size, and the prose underneath doing the work+ * that a "Project age: 1,232 days" tile never could.+ *+ * Degrades a rung at a time: without a subject it is still a dated founding,+ * without a date it renders nothing.+ */+export function OriginBlock({ data }: { data: StatsOrigin }) {+ if (!data.first_commit_at) return null;++ const author = data.first_commit_author;+ const age = data.age_days != null ? formatAgeDays(data.age_days) : null;+ const others = Math.max(0, data.contributor_count - 1);++ return (+ <section+ aria-label="Project origin"+ className="border-l-2 border-[var(--color-accent-primary)] pl-5"+ >+ <p className="font-mono text-[11px] uppercase tracking-[0.12em] text-[var(--color-text-tertiary)]">+ Root commit · {foundingDate(data.first_commit_at)}+ </p>++ {data.first_commit_subject && (+ <p className="mt-2 break-words font-mono text-[15px] font-medium leading-snug text-[var(--color-text-primary)] sm:text-lg">+ {data.first_commit_subject}+ </p>+ )}++ <p className="mt-2.5 text-sm text-[var(--color-text-secondary)]">+ {age ? (+ <>+ It started{" "}+ <strong className="font-semibold text-[var(--color-text-primary)]">{age} ago</strong>+ </>+ ) : (+ <>It started here</>+ )}+ {author && (+ <>+ , written by{" "}+ <strong className="font-semibold text-[var(--color-text-primary)]">{author}</strong>+ </>+ )}+ {". "}+ {others > 0 && (+ <span className="text-[var(--color-text-tertiary)]">+ Since then, {formatNumber(others)} more {others === 1 ? "person has" : "people have"}{" "}+ shown up.+ </span>+ )}+ </p>+ </section>+ );+}diff --git a/packages/ui/src/stats/punch-card.tsx b/packages/ui/src/stats/punch-card.tsxindex 9a484a203..f8f89987a 100644--- a/packages/ui/src/stats/punch-card.tsx+++ b/packages/ui/src/stats/punch-card.tsx@@ -2,7 +2,6 @@import * as React from "react";import type { StatsPunchCard } from "@repowise-dev/types/stats";-import { Card, CardContent, CardHeader, CardTitle } from "../ui/card";import { DEFAULT_WEEKEND_PRESET, weekendShare } from "./weekend";const WEEKDAYS = ["Mon", "Tue", "Wed", "Thu", "Fri", "Sat", "Sun"] as const;@@ -27,10 +26,21 @@ function weekdayLong(i: number): string {}/**- * Coding-rhythm heatmap: a GitHub-punch-card of commit volume by weekday x hour.- * The signature "when does this team actually code" view — nothing else on the- * stats page shows temporal shape. Cells ramp the accent color by a sqrt-scaled- * intensity, and hovering a cell reads out its weekday/hour/count live.+ * Coding-rhythm heatmap: commit volume by weekday x hour.+ *+ * The page's signature view and its focal point — nothing else in the app shows+ * temporal shape at all. Presented as an open figure rather than inside a card:+ * the heading, the live readout and the lattice are one object, and a border+ * around them only competes with the grid's own structure.+ *+ * Cells ramp the accent by a sqrt-scaled intensity so low-but-nonzero hours stay+ * legible against the peak; hovering reads out the exact cell and dims the rest.+ *+ * The clock is whatever the server resolved: `author_local` means each commit+ * was shifted by its own author's UTC offset, which is the only version of this+ * chart that means anything across timezones. The footer always names which,+ * because a heatmap that silently changes clocks is worse than one that admits+ * it hasn't got the data yet.*/export function PunchCard({data,@@ -46,6 +56,8 @@ export function PunchCard({const max = data.peak.count || 1;const weekendPct = weekendShare(data.matrix, weekendDays);+ const isLocal = data.timezone_mode === "author_local";+const readout = hover? `${weekdayLong(hover.wd)} · ${hourLabel(hover.hr)} · ${hover.count} commit${hover.count === 1 ? "" : "s"@@ -55,112 +67,118 @@ export function PunchCard({: "Commit activity by weekday and hour";return (- <Card className="h-full">- <CardHeader className="flex-row items-center justify-between gap-3 pb-3">- <CardTitle className="text-sm">Coding rhythm</CardTitle>- <span className="shrink-0 rounded-full bg-[var(--color-bg-muted)] px-2.5 py-1 text-[11px] font-medium tabular-nums text-[var(--color-text-secondary)]">+ <section aria-label="Coding rhythm" className="flex flex-col gap-4">+ <div className="flex flex-wrap items-baseline justify-between gap-x-4 gap-y-1">+ <h3 className="text-base font-semibold text-[var(--color-text-primary)]">Coding rhythm</h3>+ <span className="font-mono text-[11px] tabular-nums text-[var(--color-text-tertiary)]">{weekendPct}% on weekends</span>- </CardHeader>- <CardContent className="pt-0">- <p- className={`mb-3 text-xs transition-colors ${- hover- ? "font-medium text-[var(--color-text-primary)]"- : "text-[var(--color-text-secondary)]"- }`}- >- {readout}- </p>+ </div>- <div className="overflow-x-auto">- <div className="min-w-[420px]" onMouseLeave={() => setHover(null)}>- <div className="flex flex-col" style={{ gap: CELL_GAP }}>- {WEEKDAYS.map((day, wd) => (- <div key={day} className="flex items-center gap-1.5">- <span- className={`w-8 shrink-0 py-px text-right text-[10px] font-medium uppercase tracking-wide transition-colors ${- hover?.wd === wd- ? "text-[var(--color-accent-primary)]"- : "text-[var(--color-text-tertiary)]"- }`}- >- {day}- </span>- <div- className="grid flex-1 grid-cols-[repeat(24,minmax(0,1fr))]"- style={{ gap: CELL_GAP }}- >- {Array.from({ length: 24 }, (_, hr) => {- const count = data.matrix[wd]?.[hr] ?? 0;- // sqrt keeps low-but-nonzero hours legible against the peak.- const intensity = count > 0 ? Math.sqrt(count / max) : 0;- const isHover = hover?.wd === wd && hover?.hr === hr;- const dimmed = hover && !isHover && hover.wd !== wd && hover.hr !== hr;- return (- <div- key={hr}- onMouseEnter={() => setHover({ wd, hr, count })}- className={`aspect-square rounded-[2px] transition-all duration-100 ${- isHover- ? "scale-[1.35] ring-1 ring-[var(--color-accent-primary)]"- : ""- }`}- style={{- background:- count > 0 ? "var(--color-accent-primary)" : "var(--color-bg-muted)",- opacity: isHover- ? 1- : count > 0- ? (dimmed ? 0.5 : 1) * (0.16 + 0.84 * intensity)- : dimmed- ? 0.4- : 1,- }}- />- );- })}- </div>- </div>- ))}- </div>+ {/* The readout is the chart's title line, so it holds its height rather+ than reflowing the lattice every time the pointer moves. */}+ <p+ className={`min-h-[1.25rem] text-sm transition-colors ${+ hover+ ? "font-medium text-[var(--color-text-primary)]"+ : "text-[var(--color-text-secondary)]"+ }`}+ aria-live="polite"+ >+ {readout}+ </p>- {/* Hour axis — ticks aligned to the 24-column grid. */}- <div className="mt-1.5 flex items-center gap-1.5">- <span className="w-8 shrink-0" />- <div className="relative grid flex-1 grid-cols-[repeat(24,minmax(0,1fr))]">- {HOUR_TICKS.map(([h, label]) => (- <span- key={h}- className="col-span-1 whitespace-nowrap text-[10px] tabular-nums text-[var(--color-text-tertiary)]"- style={{ gridColumnStart: h + 1 }}- >- {label}- </span>- ))}+ <div className="overflow-x-auto">+ <div className="min-w-[420px]" onMouseLeave={() => setHover(null)}>+ <div className="flex flex-col" style={{ gap: CELL_GAP }}>+ {WEEKDAYS.map((day, wd) => (+ <div key={day} className="flex items-center gap-2">+ <span+ className={`w-8 shrink-0 py-px text-right font-mono text-[10px] uppercase tracking-wide transition-colors ${+ hover?.wd === wd+ ? "text-[var(--color-accent-primary)]"+ : "text-[var(--color-text-tertiary)]"+ }`}+ >+ {day}+ </span>+ <div+ className="grid flex-1 grid-cols-[repeat(24,minmax(0,1fr))]"+ style={{ gap: CELL_GAP }}+ >+ {Array.from({ length: 24 }, (_, hr) => {+ const count = data.matrix[wd]?.[hr] ?? 0;+ // sqrt keeps low-but-nonzero hours legible against the peak.+ const intensity = count > 0 ? Math.sqrt(count / max) : 0;+ const isHover = hover?.wd === wd && hover?.hr === hr;+ const dimmed = hover && !isHover && hover.wd !== wd && hover.hr !== hr;+ return (+ <div+ key={hr}+ onMouseEnter={() => setHover({ wd, hr, count })}+ className={`aspect-square rounded-[2px] transition-all duration-100 ${+ isHover ? "scale-[1.35] ring-1 ring-[var(--color-accent-primary)]" : ""+ }`}+ style={{+ background:+ count > 0 ? "var(--color-accent-primary)" : "var(--color-bg-muted)",+ opacity: isHover+ ? 1+ : count > 0+ ? (dimmed ? 0.5 : 1) * (0.16 + 0.84 * intensity)+ : dimmed+ ? 0.4+ : 1,+ }}+ />+ );+ })}+ </div></div>+ ))}+ </div>++ {/* Hour axis — ticks aligned to the 24-column grid. */}+ <div className="mt-2 flex items-center gap-2">+ <span className="w-8 shrink-0" />+ <div className="relative grid flex-1 grid-cols-[repeat(24,minmax(0,1fr))]">+ {HOUR_TICKS.map(([h, label]) => (+ <span+ key={h}+ className="col-span-1 whitespace-nowrap font-mono text-[10px] tabular-nums text-[var(--color-text-tertiary)]"+ style={{ gridColumnStart: h + 1 }}+ >+ {label}+ </span>+ ))}</div></div></div>+ </div>- {/* Legend — hours are bucketed in UTC (git records the commit instant,- not the author's local offset), so we say so rather than imply local- time. */}- <div className="mt-3 flex items-center justify-between gap-2 text-[10px] text-[var(--color-text-tertiary)]">- <span>Hours in UTC</span>- <div className="flex items-center gap-1.5">- <span>Less</span>- {[0.16, 0.44, 0.72, 1].map((o) => (- <span- key={o}- className="h-2.5 w-2.5 rounded-[2px]"- style={{ background: "var(--color-accent-primary)", opacity: o }}- />- ))}- <span>More</span>- </div>+ <div className="flex items-center justify-between gap-2 border-t border-[var(--color-border-default)] pt-3 text-[10px] text-[var(--color-text-tertiary)]">+ <span+ title={+ isLocal+ ? "Each commit is placed at its author's own local time, using the UTC offset git recorded with it."+ : "This index predates the commit-offset capture, so hours fall back to UTC. Run `repowise update` to switch to author-local time."+ }+ className="cursor-help font-mono uppercase tracking-[0.1em]"+ >+ {isLocal ? "Author-local time" : "Hours in UTC"}+ </span>+ <div className="flex items-center gap-1.5">+ <span>Less</span>+ {[0.16, 0.44, 0.72, 1].map((o) => (+ <span+ key={o}+ className="h-2.5 w-2.5 rounded-[2px]"+ style={{ background: "var(--color-accent-primary)", opacity: o }}+ />+ ))}+ <span>More</span></div>- </CardContent>- </Card>+ </div>+ </section>);}diff --git a/packages/ui/src/stats/superlatives-grid.tsx b/packages/ui/src/stats/records-grid.tsxsimilarity index 59%rename from packages/ui/src/stats/superlatives-grid.tsxrename to packages/ui/src/stats/records-grid.tsxindex 9c7cd455f..42287fad6 100644--- a/packages/ui/src/stats/superlatives-grid.tsx+++ b/packages/ui/src/stats/records-grid.tsx@@ -1,16 +1,19 @@import * as React from "react";import {+ Boxes,FileText,- Flame,GitCommitHorizontal,Hourglass,Network,+ Repeat,+ Ruler,+ ScrollText,+ Spline,Workflow,- Boxes,Zap,} from "lucide-react";-import type { StatsSuperlatives } from "@repowise-dev/types/stats";-import { formatDate, formatNumber, formatRelativeTimeOrNull, truncatePath } from "../lib/format";+import type { StatsRecords } from "@repowise-dev/types/stats";+import { formatNumber, formatRelativeTimeOrNull, truncatePath } from "../lib/format";interface AwardRow {key: string;@@ -20,7 +23,7 @@ interface AwardRow {detail: string;}-function buildAwards(s: StatsSuperlatives): AwardRow[] {+function buildAwards(s: StatsRecords): AwardRow[] {const rows: AwardRow[] = [];if (s.largest_file) {rows.push({@@ -31,6 +34,15 @@ function buildAwards(s: StatsSuperlatives): AwardRow[] {detail: `${formatNumber(s.largest_file.nloc)} lines`,});}+ if (s.gnarliest_file) {+ rows.push({+ key: "gnarliest",+ icon: Spline,+ title: "Gnarliest file",+ primary: truncatePath(s.gnarliest_file.path, 42),+ detail: `cyclomatic complexity ${formatNumber(s.gnarliest_file.max_ccn)}`,+ });+ }if (s.most_complex_symbol) {rows.push({key: "complex",@@ -85,15 +97,13 @@ function buildAwards(s: StatsSuperlatives): AwardRow[] {)} files — in one commit`,});}- if (s.longest_streak) {+ if (s.widest_commit) {rows.push({- key: "streak",- icon: Flame,- title: "Longest streak",- primary: `${formatNumber(s.longest_streak.days)} consecutive days`,- detail: `commits every day, ${formatDate(s.longest_streak.start)} – ${formatDate(- s.longest_streak.end,- )}`,+ key: "widest-commit",+ icon: Ruler,+ title: "Widest commit",+ primary: s.widest_commit.subject || s.widest_commit.sha.slice(0, 10),+ detail: `touched ${formatNumber(s.widest_commit.files_changed)} files at once`,});}if (s.strongest_coupling) {@@ -105,20 +115,47 @@ function buildAwards(s: StatsSuperlatives): AwardRow[] {s.strongest_coupling.b,26,)}`,- detail: `changed together ${formatNumber(s.strongest_coupling.count)}×`,+ detail: `changed together ${formatNumber(Math.round(s.strongest_coupling.count))}×`,+ });+ }+ // Import cycles have no other surface in the app — Architecture visualises the+ // graph but never detects or counts circular imports.+ if (s.largest_cycle && s.largest_cycle.files > 1) {+ rows.push({+ key: "cycle",+ icon: Repeat,+ title: "Largest import cycle",+ primary: `${formatNumber(s.largest_cycle.files)} files in a loop`,+ detail: `${formatNumber(s.largest_cycle.cycle_count)} circular cluster${+ s.largest_cycle.cycle_count === 1 ? "" : "s"+ } in total`,+ });+ }+ if (s.symbol_shape && s.symbol_shape.total > 0) {+ const shape = s.symbol_shape;+ rows.push({+ key: "documented",+ icon: ScrollText,+ title: "Documented symbols",+ primary: `${Math.round(shape.documented_pct)}% carry a docstring`,+ detail: `${formatNumber(shape.documented_count)} of ${formatNumber(shape.total)} symbols · ${Math.round(+ shape.async_pct,+ )}% async`,});}return rows;}-interface SuperlativesGridProps {- superlatives: StatsSuperlatives;-}--/** A grid of "award" cards — the biggest / oldest / most-tangled records in- * the repo. Renders only the awards that have data. */-export function SuperlativesGrid({ superlatives }: SuperlativesGridProps) {- const awards = buildAwards(superlatives);+/**+ * The repo's records — biggest, oldest, gnarliest, most tangled.+ *+ * Several of these are the only headline treatment their signal gets anywhere:+ * cyclomatic complexity is otherwise a sortable column on Code Health, and+ * import cycles have no UI at all. Renders only awards that have data, so a+ * young repo shows a short grid rather than a wall of dashes.+ */+export function RecordsGrid({ records }: { records: StatsRecords }) {+ const awards = buildAwards(records);if (awards.length === 0) return null;return (@@ -140,7 +177,10 @@ export function SuperlativesGrid({ superlatives }: SuperlativesGridProps) {>{a.primary}</p>- <p className="mt-0.5 truncate text-xs text-[var(--color-text-secondary)]" title={a.detail}>+ <p+ className="mt-0.5 truncate text-xs text-[var(--color-text-secondary)]"+ title={a.detail}+ >{a.detail}</p></div>diff --git a/packages/ui/src/stats/size-class-hero.tsx b/packages/ui/src/stats/size-class-hero.tsxindex c1fbc9db5..ff5e34942 100644--- a/packages/ui/src/stats/size-class-hero.tsx+++ b/packages/ui/src/stats/size-class-hero.tsx@@ -8,8 +8,6 @@ import {Globe2,} from "lucide-react";import type { StatsScale } from "@repowise-dev/types/stats";-import { formatCompact, formatNumber, formatLOC } from "../lib/format";-import { NLOC_HINT } from "./stat-callout";const SIZE_ICON: Record<string, React.ComponentType<{ className?: string }>> = {Seedling: Sprout,@@ -21,13 +19,6 @@ const SIZE_ICON: Record<string, React.ComponentType<{ className?: string }>> = {Megalopolis: Globe2,};-interface HeroFigure {- label: string;- value: string;- /** Longer clarification shown as a native hover tooltip. */- hint?: string;-}-interface SizeClassHeroProps {scale: StatsScale;/** Repo name, rendered as the eyebrow above the size-class label. */@@ -35,74 +26,44 @@ interface SizeClassHeroProps {}/**- * The Stats page centerpiece: a warm-washed banner that names the codebase's- * "size class" (derived from NLOC) and leads with the headline figures. Reads- * as the centerpiece through spacing, the wash, and one accent — not through- * outsized type, which would put it out of step with every other page.+ * The Stats page's opening statement: a warm-washed banner naming the+ * codebase's "size class" (derived from NLOC).+ *+ * Identity only. The headline figures used to sit here as four bordered tiles,+ * which put a boxed number immediately above the hairline ribbon carrying the+ * same four numbers — the ribbon owns them now, and the hero says the one thing+ * a count can't: what kind of place this is.*/export function SizeClassHero({ scale, repoName }: SizeClassHeroProps) {const Icon = SIZE_ICON[scale.size_class.name] ?? Building2;- const figures: HeroFigure[] = [- { label: "Lines of code", value: formatLOC(scale.total_nloc), hint: NLOC_HINT },- { label: "Files", value: formatCompact(scale.file_count) },- { label: "Symbols", value: formatCompact(scale.symbol_count) },- { label: "Languages", value: formatNumber(scale.language_count) },- ];-return (<divclassName="relative overflow-hidden rounded-2xl border border-[var(--color-border-default)] p-6 sm:p-8"style={{ background: "var(--gradient-warm-wash, var(--color-bg-surface))" }}>- {/* Side by side once there is room, but wrapping rather than letting the- figures crowd the size-class name — the figure tiles refuse to shrink,- so the row needs somewhere to break. */}- <div className="flex flex-col gap-6 lg:flex-row lg:flex-wrap lg:items-center lg:justify-between">- <div className="min-w-0 lg:basis-80 lg:grow">- {repoName && (- <p className="text-xs font-medium uppercase tracking-[0.2em] text-[var(--color-text-tertiary)]">- {repoName}+ <div className="min-w-0">+ {repoName && (+ <p className="text-xs font-medium uppercase tracking-[0.2em] text-[var(--color-text-tertiary)]">+ {repoName}+ </p>+ )}+ <div className="mt-2 flex items-center gap-3">+ <span className="flex h-12 w-12 shrink-0 items-center justify-center rounded-xl bg-[var(--color-accent-muted,var(--color-bg-elevated))] text-[var(--color-accent-primary)]">+ <Icon className="h-6 w-6" />+ </span>+ <div className="min-w-0">+ <p className="text-[11px] font-medium uppercase tracking-wider text-[var(--color-text-tertiary)]">+ This codebase is a</p>- )}- <div className="mt-2 flex items-center gap-3">- <span className="flex h-12 w-12 shrink-0 items-center justify-center rounded-xl bg-[var(--color-accent-muted,var(--color-bg-elevated))] text-[var(--color-accent-primary)]">- <Icon className="h-6 w-6" />- </span>- <div className="min-w-0">- <p className="text-[11px] font-medium uppercase tracking-wider text-[var(--color-text-tertiary)]">- This codebase is a- </p>- <h2 className="text-2xl font-bold leading-tight text-[var(--color-text-primary)] sm:text-3xl">- {scale.size_class.name}- </h2>- </div>+ <h2 className="text-2xl font-bold leading-tight text-[var(--color-text-primary)] sm:text-3xl">+ {scale.size_class.name}+ </h2></div>- <p className="mt-3 max-w-md text-sm text-[var(--color-text-secondary)]">- {scale.size_class.blurb}- </p>- </div>-- {/* Columns floor at their content width so a six-character figure- ("485.3K") is never clipped to an ellipsis; they still share the- leftover width equally, and the group refuses to be squeezed by the- blurb beside it. */}- <div className="grid grid-cols-[repeat(2,minmax(max-content,1fr))] gap-3 sm:grid-cols-[repeat(4,minmax(max-content,1fr))] lg:shrink-0">- {figures.map((f) => (- <div- key={f.label}- title={f.hint}- className={`rounded-xl border border-[var(--color-border-subtle,var(--color-border-default))] bg-[var(--color-bg-surface)]/70 px-4 py-3 backdrop-blur-sm${f.hint ? " cursor-help" : ""}`}- >- <p className="whitespace-nowrap text-2xl font-bold tabular-nums text-[var(--color-text-primary)]">- {f.value}- </p>- <p className="mt-0.5 text-[11px] font-medium uppercase tracking-wider text-[var(--color-text-tertiary)]">- {f.label}- </p>- </div>- ))}</div>+ <p className="mt-3 max-w-md text-sm text-[var(--color-text-secondary)]">+ {scale.size_class.blurb}+ </p></div></div>);diff --git a/packages/ui/src/stats/stat-ribbon.tsx b/packages/ui/src/stats/stat-ribbon.tsxnew file mode 100644index 000000000..0ca6b37a9--- /dev/null+++ b/packages/ui/src/stats/stat-ribbon.tsx@@ -0,0 +1,57 @@+import * as React from "react";++export interface RibbonStat {+ label: string;+ value: string;+ hint?: string;+}++/**+ * A row of figures separated by hairlines rather than boxed into cards.+ *+ * The stats page used to be ~25 near-identical bordered tiles, which reads as+ * box soup: every figure claims the same visual weight, so none of them lands.+ * Rules carry the same grouping at a fraction of the ink, which is why the+ * public repo pages use them for exactly this job.+ *+ * Rendered as a `<dl>` because that is what it is — labelled values, not a+ * layout grid.+ */+export function StatRibbon({ stats }: { stats: RibbonStat[] }) {+ const shown = stats.filter((s) => s.value);+ if (shown.length === 0) return null;++ return (+ <dl className="grid grid-cols-2 border-y border-[var(--color-border-default)] sm:grid-cols-3 lg:grid-cols-5">+ {shown.map((s, i) => (+ <div+ key={s.label}+ title={s.hint}+ className={[+ "px-4 py-3.5",+ s.hint ? "cursor-help" : "",+ // Hairlines between cells only — the outer edges come from the+ // wrapper's border-y, so cells never double up on the boundary.+ "border-[var(--color-border-default)]",+ i % 2 === 1 ? "border-l" : "",+ i >= 2 ? "border-t" : "",+ "sm:border-l sm:border-t-0",+ i % 3 === 0 ? "sm:border-l-0" : "",+ i >= 3 ? "sm:border-t" : "",+ "lg:border-l lg:border-t-0",+ i % 5 === 0 ? "lg:border-l-0" : "",+ ]+ .filter(Boolean)+ .join(" ")}+ >+ <dt className="font-mono text-[10px] uppercase tracking-[0.12em] text-[var(--color-text-tertiary)]">+ {s.label}+ </dt>+ <dd className="mt-1 text-xl font-semibold tabular-nums text-[var(--color-text-primary)]">+ {s.value}+ </dd>+ </div>+ ))}+ </dl>+ );+}diff --git a/packages/web/src/app/repos/[id]/overview/page.tsx b/packages/web/src/app/repos/[id]/overview/page.tsxindex c535f8625..298905799 100644--- a/packages/web/src/app/repos/[id]/overview/page.tsx+++ b/packages/web/src/app/repos/[id]/overview/page.tsx@@ -154,7 +154,7 @@ export default async function OverviewPage({ params }: Props) {repoId={id}averageDelta={stats.deltas.average_health}hotspotDelta={stats.deltas.hotspot_health}- defectAccuracy={statsHighlights?.quality.defect_accuracy ?? null}+ defectAccuracy={health.defect_accuracy ?? null}/><ContributorsStripCard repoId={id} />{/* The two short tiles sit two-up here rather than stacked indiff --git a/packages/web/src/app/repos/[id]/stats/page.tsx b/packages/web/src/app/repos/[id]/stats/page.tsxindex b88bb5df6..e90985cbb 100644--- a/packages/web/src/app/repos/[id]/stats/page.tsx+++ b/packages/web/src/app/repos/[id]/stats/page.tsx@@ -25,10 +25,10 @@ export default async function StatsPage({ params }: Props) {<PageShelltitle="By the Numbers"icon={<BarChart3 className="h-5 w-5" />}- description="A tour of everything the index knows about this codebase — scale, growth, people, and the signals only repowise computes."+ description="The things about this codebase you can't see anywhere else — how big it got, when it started, when the work actually happens, and the records it holds."maxWidth="wide">- <StatsTabs data={data} />+ <StatsTabs data={data} repoId={id} /></PageShell>);}diff --git a/packages/web/src/components/overview/stats-teaser-card.tsx b/packages/web/src/components/overview/stats-teaser-card.tsxindex eb77557ce..e858f9216 100644--- a/packages/web/src/components/overview/stats-teaser-card.tsx+++ b/packages/web/src/components/overview/stats-teaser-card.tsx@@ -1,9 +1,9 @@import Link from "next/link";-import { BarChart3, ArrowRight, Bot } from "lucide-react";+import { BarChart3, ArrowRight, Clock } from "lucide-react";import type { StatsHighlights } from "@repowise-dev/types/stats";import { Card } from "@repowise-dev/ui/ui/card";import { formatLOC, formatNumber, formatAgeDays } from "@repowise-dev/ui/lib/format";-import { NLOC_HINT, AGENT_PCT_HINT } from "@repowise-dev/ui/stats";+import { NLOC_HINT } from "@repowise-dev/ui/stats";interface StatsTeaserCardProps {repoId: string;@@ -13,14 +13,14 @@ interface StatsTeaserCardProps {/** Compact overview teaser that headlines the "size class" + a couple of* marquee figures and links into the full Stats page. */export function StatsTeaserCard({ repoId, data }: StatsTeaserCardProps) {- const { scale, activity } = data;+ const { scale, origin, rhythm } = data;const figures: { label: string; value: string; hint?: string }[] = [{ label: "Lines of code", value: formatLOC(scale.total_nloc), hint: NLOC_HINT },- { label: "Commits", value: formatNumber(activity.total_commits) },+ { label: "Commits", value: formatNumber(origin.total_commits) },{label: "Age",- value: activity.age_days != null ? formatAgeDays(activity.age_days) : "—",+ value: origin.age_days != null ? formatAgeDays(origin.age_days) : "—",},];@@ -60,13 +60,10 @@ export function StatsTeaserCard({ repoId, data }: StatsTeaserCardProps) {))}</div>- {activity.total_commits > 0 && (- <p- title={AGENT_PCT_HINT}- className="mt-3 flex cursor-help items-center gap-1.5 text-xs text-[var(--color-text-secondary)]"- >- <Bot className="h-3.5 w-3.5 text-[var(--color-info)]" />- {activity.agent_pct}% of commits carry a verifiable agent signature+ {rhythm.longest_streak && (+ <p className="mt-3 flex items-center gap-1.5 text-xs text-[var(--color-text-secondary)]">+ <Clock className="h-3.5 w-3.5 text-[var(--color-accent-primary)]" />+ Longest streak: {formatNumber(rhythm.longest_streak.days)} consecutive days</p>)}</Link>diff --git a/packages/web/src/components/stats/architecture-tab.tsx b/packages/web/src/components/stats/architecture-tab.tsxdeleted file mode 100644index 89b650301..000000000--- a/packages/web/src/components/stats/architecture-tab.tsx+++ /dev/null@@ -1,79 +0,0 @@-import { Boxes, Lightbulb, Waypoints, FileCode, Repeat, Network } from "lucide-react";-import type { StatsHighlights } from "@repowise-dev/types/stats";-import { StatCallout, SuperlativesGrid } from "@repowise-dev/ui/stats";-import { LanguageDonut } from "@repowise-dev/ui/dashboard/language-donut";-import { formatNumber } from "@repowise-dev/ui/lib/format";--export function ArchitectureTab({ data }: { data: StatsHighlights }) {- const { scale, knowledge, quality, superlatives } = data;- const graph = data.graph;-- const langDistribution: Record<string, number> = {};- for (const l of scale.languages) langDistribution[l.language] = l.file_count;-- return (- <div className="space-y-5">- <div className="grid grid-cols-1 gap-3 sm:grid-cols-2 lg:grid-cols-3">- <StatCallout- label="Modules"- value={formatNumber(scale.module_count)}- icon={<Boxes className="h-4 w-4" />}- sub={`${formatNumber(scale.file_count)} files`}- />- <StatCallout- label="Entry points"- value={formatNumber(scale.entry_point_count)}- icon={<Waypoints className="h-4 w-4" />}- sub="reachable roots in the graph"- />- <StatCallout- label="Symbols"- value={formatNumber(scale.symbol_count)}- icon={<FileCode className="h-4 w-4" />}- sub={`${formatNumber(scale.language_count)} languages · ${Math.round(- quality.doc_coverage_pct,- )}% documented`}- />- <StatCallout- label="Decisions"- value={formatNumber(knowledge.decision_count)}- icon={<Lightbulb className="h-4 w-4" />}- sub={`${formatNumber(knowledge.active_decision_count)} active`}- />- {graph && (- <StatCallout- label="Dependency cycles"- value={formatNumber(graph.cycle_clusters)}- tone={graph.cycle_clusters > 0 ? "warning" : "success"}- icon={<Repeat className="h-4 w-4" />}- sub={- graph.cycle_clusters > 0- ? `largest spans ${formatNumber(graph.largest_cycle)} files · ${formatNumber(- graph.files_in_cycles,- )} caught in a loop`- : "no circular imports — clean layering"- }- />- )}- {graph && (- <StatCallout- label="Neighborhoods"- value={formatNumber(graph.community_count)}- icon={<Network className="h-4 w-4" />}- sub="natural clusters in the import graph"- />- )}- </div>-- <div className="grid grid-cols-1 gap-4 lg:grid-cols-3">- <LanguageDonut distribution={langDistribution} />- <div className="lg:col-span-2">- <h3 className="mb-2 text-sm font-semibold text-[var(--color-text-primary)]">- Structural records- </h3>- <SuperlativesGrid superlatives={superlatives} />- </div>- </div>- </div>- );-}diff --git a/packages/web/src/components/stats/by-the-numbers-tab.tsx b/packages/web/src/components/stats/by-the-numbers-tab.tsxindex a807246fa..b933f2ea7 100644--- a/packages/web/src/components/stats/by-the-numbers-tab.tsx+++ b/packages/web/src/components/stats/by-the-numbers-tab.tsx@@ -1,261 +1,57 @@-import Link from "next/link";-import {- Bot,- CalendarClock,- Users,- ShieldCheck,- Trash2,- HeartPulse,- Package,- TrendingUp,- TrendingDown,- Sparkles,-} from "lucide-react";import type { StatsHighlights } from "@repowise-dev/types/stats";import {+ ChurnLedger,+ NLOC_HINT,+ OriginBlock,+ RecordsGrid,SizeClassHero,- StatCallout,- SuperlativesGrid,- PunchCard,- AGENT_PCT_HINT,+ StatRibbon,+ type RibbonStat,} from "@repowise-dev/ui/stats";-import {- formatNumber,- formatAgeDays,- formatDate,- formatTokens,- formatCost,-} from "@repowise-dev/ui/lib/format";-import { useWeekendDays } from "@/lib/hooks/use-weekend";--const ECOSYSTEM_NAMES: Record<string, string> = {- npm: "npm",- pypi: "PyPI",- go: "Go",- cargo: "crates.io",- nuget: "NuGet",-};-+import { formatLOC, formatNumber } from "@repowise-dev/ui/lib/format";++/**+ * Tab 1 — what this repo *is*: scale, origin, lifetime churn, records.+ *+ * Deliberately carries no health scores, commit volume or dependency counts:+ * each of those has a page that owns the subject and shows it with filtering+ * and drill-down this page could never justify. What is left is the repo's+ * fingerprint, which nothing else draws.+ */export function ByTheNumbersTab({ data }: { data: StatsHighlights }) {- const { scale, activity, quality, people, superlatives } = data;- const da = quality.defect_accuracy;- const deps = data.dependencies;- const build = data.build;- const vel = activity.velocity;- const weekendDays = useWeekendDays();+ const { scale, origin, churn, records, rhythm } = data;- // Momentum: recent-vs-prior commit change, or a plain recent count when there- // is no prior window to compare against (young repo).- const rising = vel?.pct_change != null && vel.pct_change >= 0;- const momentum =- vel?.pct_change != null- ? `${rising ? "+" : ""}${vel.pct_change}%`- : formatNumber(vel?.recent_90d ?? 0);+ const ribbon: RibbonStat[] = [+ { label: "Lines of code", value: formatLOC(scale.total_nloc), hint: NLOC_HINT },+ { label: "Files", value: formatNumber(scale.file_count) },+ { label: "Symbols", value: formatNumber(scale.symbol_count) },+ { label: "Modules", value: formatNumber(scale.module_count) },+ { label: "Languages", value: formatNumber(scale.language_count) },+ ];return (- <div className="space-y-5">+ <div className="flex flex-col gap-8"><SizeClassHero scale={scale} repoName={data.repo.name} />- <div className="grid grid-cols-1 gap-3 sm:grid-cols-2 lg:grid-cols-4">- <StatCallout- label="AI authorship"- value={`${activity.agent_pct}%`}- tone="info"- icon={<Bot className="h-4 w-4" />}- sub={`${formatNumber(activity.agent_commits)} of ${formatNumber(- activity.total_commits,- )} commits carry a verifiable agent signature`}- hint={AGENT_PCT_HINT}- />- <StatCallout- label="Project age"- value={activity.age_days != null ? formatAgeDays(activity.age_days) : "—"}- icon={<CalendarClock className="h-4 w-4" />}- sub={- activity.first_commit_at- ? `since ${formatDate(activity.first_commit_at)}${- activity.first_commit_author ? ` · started by ${activity.first_commit_author}` : ""- }`- : "first commit date unknown"- }- />- <StatCallout- label="Contributors"- value={formatNumber(activity.contributor_count)}- icon={<Users className="h-4 w-4" />}- sub={`${formatNumber(people.owner_count)} hold primary ownership of a file`}- />- {da ? (- <StatCallout- label="Score finds bugs"- value={da.lift != null ? `${da.lift}×` : `${Math.round(da.precision * 100)}%`}- tone="success"- icon={<ShieldCheck className="h-4 w-4" />}- sub={`${da.hits}/${da.k} least-healthy files later got a bug fix — ${- da.lift != null ? `${da.lift}× the base rate` : "above base rate"- }`}- />- ) : (- <StatCallout- label="Defect risk"- value={quality.average_health != null ? `${quality.average_health.toFixed(1)}/10` : "—"}- tone="accent"- icon={<HeartPulse className="h-4 w-4" />}- sub="average health across all files"- />- )}- </div>+ <OriginBlock data={origin} />- <div className="grid grid-cols-1 gap-4 lg:grid-cols-3">- <div className="lg:col-span-2">- <PunchCard data={activity.punch_card} weekendDays={weekendDays} />- </div>- <div className="flex flex-col gap-4">- {/* Momentum — recent vs prior 90-day commit volume, with a mini- comparison bar so the card carries its own weight. */}- <div className="flex flex-1 flex-col justify-between rounded-xl border border-[var(--color-border-default)] bg-[var(--color-bg-surface)] p-4">- <div className="flex items-center justify-between gap-2">- <p className="text-[11px] font-medium uppercase tracking-wider text-[var(--color-text-tertiary)]">- Momentum- </p>- <span- className={- vel?.pct_change == null- ? "text-[var(--color-text-tertiary)]"- : rising- ? "text-[var(--color-success)]"- : "text-[var(--color-warning)]"- }- >- {vel?.pct_change == null ? null : rising ? (- <TrendingUp className="h-4 w-4" />- ) : (- <TrendingDown className="h-4 w-4" />- )}- </span>- </div>- <p- className={`mt-1.5 text-2xl font-bold leading-none tabular-nums ${- vel?.pct_change == null- ? "text-[var(--color-text-primary)]"- : rising- ? "text-[var(--color-success)]"- : "text-[var(--color-warning)]"- }`}- >- {momentum}- </p>- {vel && vel.pct_change != null ? (- <div className="mt-4 space-y-2">- {(- [- { label: "Prev 90d", value: vel.prior_90d, accent: false },- { label: "Last 90d", value: vel.recent_90d, accent: true },- ] as const- ).map((r) => {- const barMax = Math.max(vel.recent_90d, vel.prior_90d, 1);- return (- <div key={r.label} className="flex items-center gap-2 text-[11px]">- <span className="w-14 shrink-0 text-[var(--color-text-tertiary)]">- {r.label}- </span>- <div className="h-1.5 flex-1 overflow-hidden rounded-full bg-[var(--color-bg-muted)]">- <div- className="h-full rounded-full"- style={{- width: `${(r.value / barMax) * 100}%`,- background: r.accent- ? "var(--color-accent-primary)"- : "var(--color-text-tertiary)",- }}- />- </div>- <span className="w-8 shrink-0 text-right tabular-nums text-[var(--color-text-secondary)]">- {formatNumber(r.value)}- </span>- </div>- );- })}- </div>- ) : (- <p className="mt-2 text-xs text-[var(--color-text-secondary)]">- commits in the last 90 days- </p>- )}- </div>+ <StatRibbon stats={ribbon} />++ {churn && <ChurnLedger data={churn} />}- {/* This wiki, by the numbers — the knowledge base bragging about- itself. Spread top-to-bottom so it fills the column. */}- {build && build.page_count > 0 && (- <div- className="flex flex-1 flex-col justify-between overflow-hidden rounded-xl border border-[var(--color-border-default)] p-4"- style={{ background: "var(--gradient-warm-wash, var(--color-bg-surface))" }}- >- <div className="flex items-center gap-2">- <Sparkles className="h-4 w-4 text-[var(--color-accent-primary)]" />- <p className="text-[11px] font-medium uppercase tracking-wider text-[var(--color-text-tertiary)]">- This wiki, by the numbers- </p>- </div>- <div className="my-3 grid grid-cols-3 gap-2">- {[- { v: formatNumber(build.page_count), l: "pages" },- { v: formatTokens(build.total_tokens), l: "tokens" },- { v: formatCost(build.cost_usd), l: "to build" },- ].map((s) => (- <div key={s.l}>- <p className="text-lg font-semibold tabular-nums text-[var(--color-text-primary)]">- {s.v}- </p>- <p className="mt-0.5 text-[11px] text-[var(--color-text-tertiary)]">{s.l}</p>- </div>- ))}- </div>- <p className="text-[11px] text-[var(--color-text-secondary)]">- distilled from your codebase in {formatNumber(build.llm_operations)} model calls- </p>- </div>+ <section className="flex flex-col gap-4">+ <div className="flex flex-wrap items-baseline justify-between gap-x-4 gap-y-1">+ <h3 className="text-base font-semibold text-[var(--color-text-primary)]">+ Records & superlatives+ </h3>+ {rhythm.longest_streak && (+ <span className="font-mono text-[11px] tabular-nums text-[var(--color-text-tertiary)]">+ longest streak · {formatNumber(rhythm.longest_streak.days)} days+ </span>)}</div>- </div>-- <div className="grid grid-cols-1 gap-3 sm:grid-cols-2 lg:grid-cols-3">- <StatCallout- label="Knowledge captured"- value={formatNumber(data.knowledge.decision_count)}- sub={`architectural decisions · ${formatNumber(quality.page_count)} wiki pages`}- />- <StatCallout- label="Recoverable lines"- value={formatNumber(quality.dead_code.deletable_lines)}- tone="warning"- icon={<Trash2 className="h-4 w-4" />}- sub={`safe to delete across ${formatNumber(- quality.dead_code.total_findings,- )} dead-code findings`}- />- {deps && deps.total > 0 && (- <StatCallout- label="Dependencies"- value={formatNumber(deps.total)}- icon={<Package className="h-4 w-4" />}- href={`/repos/${data.repo.id}/architecture?view=deps`}- LinkComponent={Link}- sub={`standing on ${formatNumber(deps.runtime)} runtime + ${formatNumber(- deps.dev,- )} dev shoulders across ${deps.ecosystems- .map((e) => ECOSYSTEM_NAMES[e.name] ?? e.name)- .join(", ")}`}- />- )}- </div>-- <div>- <h3 className="mb-2 text-sm font-semibold text-[var(--color-text-primary)]">- Records & superlatives- </h3>- <SuperlativesGrid superlatives={superlatives} />- </div>+ <RecordsGrid records={records} />+ </section></div>);}diff --git a/packages/web/src/components/stats/growth-tab.tsx b/packages/web/src/components/stats/growth-tab.tsxdeleted file mode 100644index 8bfd63769..000000000--- a/packages/web/src/components/stats/growth-tab.tsx+++ /dev/null@@ -1,119 +0,0 @@-import type { StatsHighlights } from "@repowise-dev/types/stats";-import { ActivityTrendChart, StatCallout, AGENT_PCT_HINT } from "@repowise-dev/ui/stats";-import { Card, CardContent, CardHeader, CardTitle } from "@repowise-dev/ui/ui/card";-import { formatNumber, formatDate } from "@repowise-dev/ui/lib/format";--function monthLabel(month: string): string {- // "2026-06" → "Jun 2026"- const [y, m] = month.split("-");- const d = new Date(Number(y), Number(m) - 1, 1);- return new Intl.DateTimeFormat("en-US", { month: "short", year: "numeric" }).format(d);-}--const RISK_BANDS = [- { key: "low", label: "Low", color: "var(--color-success)" },- { key: "moderate", label: "Moderate", color: "var(--color-warning)" },- { key: "high", label: "High", color: "var(--color-error)" },-] as const;--export function GrowthTab({ data }: { data: StatsHighlights }) {- const { activity } = data;- const risk = activity.change_risk_mix;- const riskTotal = risk.low + risk.moderate + risk.high;-- return (- <div className="space-y-5">- <div className="grid grid-cols-1 gap-3 sm:grid-cols-2 lg:grid-cols-4">- <StatCallout- label="Total commits"- value={formatNumber(activity.total_commits)}- sub={- activity.first_commit_at && activity.last_commit_at- ? `${formatDate(activity.first_commit_at)} → ${formatDate(activity.last_commit_at)}`- : "across the indexed window"- }- />- <StatCallout- label="Busiest month"- value={activity.busiest_month ? monthLabel(activity.busiest_month.month) : "—"}- sub={- activity.busiest_month- ? `${formatNumber(activity.busiest_month.total)} commits`- : "no monthly data"- }- />- <StatCallout- label="Fix commits"- value={`${activity.fix_pct}%`}- tone="warning"- sub={`${formatNumber(activity.fix_commits)} commits classified as fixes`}- />- <StatCallout- label="Agent commits"- value={`${activity.agent_pct}%`}- tone="info"- sub={`${formatNumber(activity.agent_commits)} commits with verifiable agent signatures`}- hint={AGENT_PCT_HINT}- />- </div>-- {riskTotal > 0 && (- <Card>- <CardHeader className="flex-row items-center justify-between gap-3 pb-3">- <CardTitle className="text-sm">Change-risk mix</CardTitle>- <span className="text-[11px] text-[var(--color-text-tertiary)]">- calibrated per-commit risk across {formatNumber(riskTotal)} commits- </span>- </CardHeader>- <CardContent className="pt-0">- <div className="flex h-3 w-full overflow-hidden rounded-full">- {RISK_BANDS.map((b) => {- const pct = (risk[b.key] / riskTotal) * 100;- return pct > 0 ? (- <div- key={b.key}- style={{ width: `${pct}%`, background: b.color }}- title={`${b.label}: ${risk[b.key]} (${Math.round(pct)}%)`}- />- ) : null;- })}- </div>- <div className="mt-3 grid grid-cols-3 gap-2 text-xs">- {RISK_BANDS.map((b) => (- <div key={b.key} className="flex items-center gap-1.5">- <span className="h-2 w-2 rounded-full" style={{ background: b.color }} />- <span className="text-[var(--color-text-secondary)]">{b.label}</span>- <span className="tabular-nums text-[var(--color-text-tertiary)]">- {formatNumber(risk[b.key])}- </span>- </div>- ))}- </div>- </CardContent>- </Card>- )}-- <ActivityTrendChart monthly={activity.monthly} />-- {activity.agent_names.length > 0 && (- <Card>- <CardHeader className="pb-2">- <CardTitle className="text-sm">Agents at work</CardTitle>- </CardHeader>- <CardContent className="pt-0">- <ul className="divide-y divide-[var(--color-border-subtle,var(--color-border-default))]">- {activity.agent_names.map((a) => (- <li key={a.name} className="flex items-center justify-between py-2 text-sm">- <span className="truncate text-[var(--color-text-primary)]">{a.name}</span>- <span className="tabular-nums text-[var(--color-text-secondary)]">- {formatNumber(a.count)} commits- </span>- </li>- ))}- </ul>- </CardContent>- </Card>- )}- </div>- );-}diff --git a/packages/web/src/components/stats/people-tab.tsx b/packages/web/src/components/stats/people-tab.tsxindex 2275ecdc0..9ba0b6082 100644--- a/packages/web/src/components/stats/people-tab.tsx+++ b/packages/web/src/components/stats/people-tab.tsx@@ -1,84 +1,73 @@-import { Users, UserCheck, AlertTriangle, Truck } from "lucide-react";+import Link from "next/link";+import { AlertTriangle, Users } from "lucide-react";import type { StatsHighlights } from "@repowise-dev/types/stats";-import { StatCallout } from "@repowise-dev/ui/stats";-import { Card, CardContent, CardHeader, CardTitle } from "@repowise-dev/ui/ui/card";+import { ArrivalsTimeline, ChronotypeList, StatCallout } from "@repowise-dev/ui/stats";import { formatNumber } from "@repowise-dev/ui/lib/format";-export function PeopleTab({ data }: { data: StatsHighlights }) {- const { people, activity } = data;- const maxFiles = people.top_owners[0]?.file_count ?? 0;- const truck = people.truck_factor;+/**+ * Tab 3 — the human shape of the repo.+ *+ * Repo-level concentration and human-interest angles only. Per-person ownership+ * share, hotspots owned and dead-code burden all belong to the Contributors+ * page, which opens with a distribution bar and links straight into each+ * profile — a "top owners" list here would just be a worse copy of it, so this+ * tab points there instead.+ */+export function PeopleTab({ data, repoId }: { data: StatsHighlights; repoId: string }) {+ const { people } = data;+ const utcMode = data.rhythm.punch_card.timezone_mode === "utc";return (- <div className="space-y-5">- <div className="grid grid-cols-1 gap-3 sm:grid-cols-2 lg:grid-cols-5">- {truck != null && (- <StatCallout- label="Truck factor"- value={truck}- tone={truck <= 2 ? "warning" : "default"}- icon={<Truck className="h-4 w-4" />}- sub={- truck === 1- ? "one person owns most of the code"- : `lose ${truck} people and >50% of files lose their owner`- }- />- )}+ <div className="flex flex-col gap-8">+ <div className="grid grid-cols-1 gap-3 sm:grid-cols-2 lg:grid-cols-4"><StatCalloutlabel="Contributors"- value={formatNumber(activity.contributor_count)}+ value={formatNumber(people.contributor_count)}icon={<Users className="h-4 w-4" />}- sub="distinct commit authors"- />- <StatCallout- label="File owners"- value={formatNumber(people.owner_count)}- icon={<UserCheck className="h-4 w-4" />}- sub="hold primary ownership of ≥1 file"+ sub={`${formatNumber(people.owner_count)} own at least one file`}+ href={`/repos/${repoId}/owners`}+ LinkComponent={Link}/>+ {people.truck_factor != null && (+ <StatCallout+ label="Truck factor"+ value={formatNumber(people.truck_factor)}+ icon={<AlertTriangle className="h-4 w-4" />}+ tone={people.truck_factor <= 2 ? "warning" : "default"}+ sub="people who together own most of the code"+ hint="The fewest primary owners who between them hold more than half the owned files. A factor of 1 means one person owns most of the codebase."+ />+ )}<StatCalloutlabel="Single-owner files"value={formatNumber(people.single_owner_files)}tone="warning"- icon={<AlertTriangle className="h-4 w-4" />}- sub="bus factor of 1 — knowledge risk"+ sub="only one person has ever really touched them"/><StatCalloutlabel="Knowledge silos"value={formatNumber(people.silo_count)}- tone="warning"- sub="modules >80% owned by one person"+ tone={people.silo_count > 0 ? "warning" : "default"}+ sub="modules where one person owns over 80%"/></div>- {people.top_owners.length > 0 && (- <Card>- <CardHeader className="pb-2">- <CardTitle className="text-sm">Top contributors by files owned</CardTitle>- </CardHeader>- <CardContent className="pt-0">- <ul className="space-y-2.5">- {people.top_owners.map((o) => (- <li key={o.name} className="space-y-1">- <div className="flex items-center justify-between text-sm">- <span className="truncate text-[var(--color-text-primary)]">{o.name}</span>- <span className="tabular-nums text-[var(--color-text-secondary)]">- {formatNumber(o.file_count)} files · {Math.round(o.pct * 100)}%- </span>- </div>- <div className="h-1.5 w-full overflow-hidden rounded-full bg-[var(--color-bg-muted)]">- <div- className="h-full rounded-full bg-[var(--color-accent-primary)]"- style={{ width: `${maxFiles ? (o.file_count / maxFiles) * 100 : 0}%` }}- />- </div>- </li>- ))}- </ul>- </CardContent>- </Card>+ {people.chronotypes.length > 0 ? (+ <ChronotypeList people={people.chronotypes} />+ ) : (+ utcMode && (+ <p className="rounded-xl border border-[var(--color-border-default)] bg-[var(--color-bg-surface)] p-4 text-sm text-[var(--color-text-secondary)]">+ Commit-hour habits need each commit's local timezone, which this index was built+ before repowise captured. Run{" "}+ <code className="rounded bg-[var(--color-bg-muted)] px-1.5 py-0.5 font-mono text-xs">+ repowise update+ </code>{" "}+ to backfill it — no re-index required.+ </p>+ ))}++ <ArrivalsTimeline arrivals={people.arrivals} /></div>);}diff --git a/packages/web/src/components/stats/quality-tab.tsx b/packages/web/src/components/stats/quality-tab.tsxdeleted file mode 100644index b200fa12c..000000000--- a/packages/web/src/components/stats/quality-tab.tsx+++ /dev/null@@ -1,119 +0,0 @@-import type { StatsHighlights } from "@repowise-dev/types/stats";-import { StatCallout } from "@repowise-dev/ui/stats";-import { LanguageDonut } from "@repowise-dev/ui/dashboard/language-donut";-import { Card, CardContent, CardHeader, CardTitle } from "@repowise-dev/ui/ui/card";-import { formatNumber, truncatePath } from "@repowise-dev/ui/lib/format";--const BAND_COLOR: Record<string, string> = {- healthy: "var(--color-success)",- warning: "var(--color-warning)",- alert: "var(--color-error)",-};--function score(v: number | null): string {- return v != null ? `${v.toFixed(1)}/10` : "—";-}--export function QualityTab({ data }: { data: StatsHighlights }) {- const { quality, scale } = data;- const dist = quality.distribution;-- const langDistribution: Record<string, number> = {};- for (const l of scale.languages) langDistribution[l.language] = l.file_count;-- return (- <div className="space-y-5">- <div className="grid grid-cols-1 gap-3 sm:grid-cols-2 lg:grid-cols-4">- <StatCallout- label="Defect risk"- value={score(quality.average_health)}- tone="accent"- sub="bug-predictive headline score"- />- <StatCallout- label="Maintainability"- value={score(quality.maintainability_average)}- sub="cohesion, brain methods, DRY"- />- <StatCallout- label="Performance"- value={score(quality.performance_average)}- sub="I/O-in-loop / N+1 risk"- />- <StatCallout- label="Test coverage"- value={quality.test_coverage_pct != null ? `${quality.test_coverage_pct}%` : "—"}- sub="files with a matching test"- />- </div>-- <div className="grid grid-cols-1 gap-4 lg:grid-cols-2">- {dist && (- <Card>- <CardHeader className="pb-2">- <CardTitle className="text-sm">Health distribution</CardTitle>- </CardHeader>- <CardContent className="pt-0">- <div className="flex h-3 w-full overflow-hidden rounded-full">- {(["healthy", "warning", "alert"] as const).map((b) =>- dist.bands[b].pct > 0 ? (- <div- key={b}- style={{ width: `${dist.bands[b].pct}%`, background: BAND_COLOR[b] }}- title={`${b}: ${dist.bands[b].pct}%`}- />- ) : null,- )}- </div>- <div className="mt-3 grid grid-cols-3 gap-2 text-xs">- {(["healthy", "warning", "alert"] as const).map((b) => (- <div key={b} className="flex items-center gap-1.5">- <span- className="h-2 w-2 rounded-full"- style={{ background: BAND_COLOR[b] }}- />- <span className="capitalize text-[var(--color-text-secondary)]">{b}</span>- <span className="tabular-nums text-[var(--color-text-tertiary)]">- {dist.bands[b].files}- </span>- </div>- ))}- </div>- </CardContent>- </Card>- )}-- <LanguageDonut distribution={langDistribution} />- </div>-- <div className="grid grid-cols-1 gap-3 sm:grid-cols-2 lg:grid-cols-4">- <StatCallout- label="Open findings"- value={formatNumber(quality.open_findings)}- sub={`${quality.severity_breakdown.critical} critical · ${quality.severity_breakdown.high} high`}- />- <StatCallout- label="Dead code"- value={formatNumber(quality.dead_code.deletable_lines)}- tone="warning"- sub={`recoverable lines · ${formatNumber(quality.dead_code.total_findings)} findings`}- />- <StatCallout- label="Doc coverage"- value={`${Math.round(quality.doc_coverage_pct)}%`}- sub={`${formatNumber(quality.page_count)} wiki pages`}- />- <StatCallout- label="Lowest-scoring file"- value={quality.worst_performer_score != null ? score(quality.worst_performer_score) : "—"}- tone="warning"- sub={- quality.worst_performer_path- ? truncatePath(quality.worst_performer_path, 30)- : "no data"- }- />- </div>- </div>- );-}diff --git a/packages/web/src/components/stats/rhythm-tab.tsx b/packages/web/src/components/stats/rhythm-tab.tsxnew file mode 100644index 000000000..a99f44bb0--- /dev/null+++ b/packages/web/src/components/stats/rhythm-tab.tsx@@ -0,0 +1,106 @@+"use client";++import { CalendarDays, Flame, Snowflake, TrendingDown, TrendingUp } from "lucide-react";+import type { StatsHighlights } from "@repowise-dev/types/stats";+import { PunchCard, StatCallout } from "@repowise-dev/ui/stats";+import { formatDate, formatNumber } from "@repowise-dev/ui/lib/format";+import { useWeekendDays } from "@/lib/hooks/use-weekend";++function monthLabel(key: string): string {+ const [y, m] = key.split("-");+ // Constructed from parts rather than parsed from a string so the value is+ // never nudged across a month boundary by the browser's offset. Day is+ // pinned to the 1st and only month/year are ever displayed.+ return new Date(Number(y), Number(m) - 1, 1).toLocaleDateString(undefined, {+ month: "short",+ year: "numeric",+ });+}++/**+ * Tab 2 — when the work happens.+ *+ * The one subject with no competing page: Commits ranks individual commits by+ * risk and charts their categories over calendar time, but nothing anywhere+ * else in the app has a clock. Time-of-day, streaks, peaks, and how fast code+ * goes cold all live here.+ */+export function RhythmTab({ data }: { data: StatsHighlights }) {+ const { rhythm } = data;+ const weekendDays = useWeekendDays();+ const vel = rhythm.velocity;++ const rising = vel?.pct_change != null && vel.pct_change >= 0;+ const momentum =+ vel?.pct_change != null+ ? `${rising ? "+" : ""}${vel.pct_change}%`+ : formatNumber(vel?.recent_90d ?? 0);++ return (+ <div className="flex flex-col gap-8">+ <PunchCard data={rhythm.punch_card} weekendDays={weekendDays} />++ <div className="grid grid-cols-1 gap-3 sm:grid-cols-2 lg:grid-cols-4">+ <StatCallout+ label="Momentum"+ value={momentum}+ icon={rising ? <TrendingUp className="h-4 w-4" /> : <TrendingDown className="h-4 w-4" />}+ tone={vel?.pct_change == null ? "default" : rising ? "success" : "warning"}+ sub={+ vel?.pct_change != null+ ? `${formatNumber(vel.recent_90d)} commits in 90d vs ${formatNumber(vel.prior_90d)} before`+ : `${formatNumber(vel?.recent_90d ?? 0)} commits in the last 90 days`+ }+ hint="The 90 days ending at the newest commit, against the 90 before it. Anchored to the latest commit rather than today, so a stale index doesn't read as a slowdown."+ />++ {rhythm.longest_streak && (+ <StatCallout+ label="Longest streak"+ value={`${formatNumber(rhythm.longest_streak.days)} days`}+ icon={<Flame className="h-4 w-4" />}+ tone="accent"+ sub={`${formatDate(rhythm.longest_streak.start)} – ${formatDate(+ rhythm.longest_streak.end,+ )}`}+ />+ )}++ {rhythm.busiest_day && (+ <StatCallout+ label="Busiest day"+ value={formatNumber(rhythm.busiest_day.commits)}+ icon={<CalendarDays className="h-4 w-4" />}+ sub={`commits on ${formatDate(rhythm.busiest_day.date)}`}+ />+ )}++ {rhythm.code_half_life_days != null && (+ <StatCallout+ label="Code half-life"+ value={`${formatNumber(rhythm.code_half_life_days)} days`}+ icon={<Snowflake className="h-4 w-4" />}+ tone="info"+ sub="half the files haven't been touched in longer than this"+ hint="Median time since each file was last changed, measured from the newest commit."+ />+ )}+ </div>++ <div className="grid grid-cols-1 gap-3 sm:grid-cols-2">+ {rhythm.busiest_month && (+ <StatCallout+ label="Busiest month"+ value={monthLabel(rhythm.busiest_month.month)}+ sub={`${formatNumber(rhythm.busiest_month.total)} commits`}+ />+ )}+ <StatCallout+ label="Active days"+ value={formatNumber(rhythm.active_days)}+ sub="distinct days with at least one commit"+ />+ </div>+ </div>+ );+}diff --git a/packages/web/src/components/stats/stats-tabs.tsx b/packages/web/src/components/stats/stats-tabs.tsxindex f1a7b424a..452f77ead 100644--- a/packages/web/src/components/stats/stats-tabs.tsx+++ b/packages/web/src/components/stats/stats-tabs.tsx@@ -5,51 +5,54 @@ import { Tabs, TabsContent, TabsList, TabsTrigger } from "@repowise-dev/ui/ui/taimport { ErrorBoundary } from "@repowise-dev/ui/shared";import type { StatsHighlights } from "@repowise-dev/types/stats";import { ByTheNumbersTab } from "./by-the-numbers-tab";-import { GrowthTab } from "./growth-tab";+import { RhythmTab } from "./rhythm-tab";import { PeopleTab } from "./people-tab";-import { QualityTab } from "./quality-tab";-import { ArchitectureTab } from "./architecture-tab";-const TAB_VALUES = ["numbers", "growth", "people", "quality", "architecture"] as const;+/**+ * Three tabs, down from five.+ *+ * "Code & Quality" and "Architecture" were removed outright: every figure on+ * them had a richer counterpart on Code Health, Architecture or Knowledge+ * Graph, and "Growth & Activity" was almost entirely the Commits page. The+ * legacy tab values still resolve so an old bookmark lands somewhere sensible+ * rather than on an empty page.+ */+const TAB_VALUES = ["numbers", "rhythm", "people"] as const;+type TabValue = (typeof TAB_VALUES)[number];-export function StatsTabs({ data }: { data: StatsHighlights }) {- const [tab, setTab] = useQueryState(+const LEGACY_TABS: Record<string, TabValue> = {+ growth: "rhythm",+ quality: "numbers",+ architecture: "numbers",+};++export function StatsTabs({ data, repoId }: { data: StatsHighlights; repoId: string }) {+ const [raw, setTab] = useQueryState("tab",- parseAsStringLiteral(TAB_VALUES).withDefault("numbers"),+ parseAsStringLiteral([...TAB_VALUES, ...Object.keys(LEGACY_TABS)]).withDefault("numbers"),);+ const tab = (LEGACY_TABS[raw] ?? raw) as TabValue;return (- <Tabs value={tab} onValueChange={(v) => setTab(v as (typeof TAB_VALUES)[number])}>+ <Tabs value={tab} onValueChange={(v) => setTab(v as TabValue)}><TabsList><TabsTrigger value="numbers">By the Numbers</TabsTrigger>- <TabsTrigger value="growth">Growth & Activity</TabsTrigger>+ <TabsTrigger value="rhythm">Rhythm</TabsTrigger><TabsTrigger value="people">People</TabsTrigger>- <TabsTrigger value="quality">Code & Quality</TabsTrigger>- <TabsTrigger value="architecture">Architecture</TabsTrigger></TabsList><TabsContent value="numbers" className="mt-4"><ErrorBoundary title="Couldn't load stats"><ByTheNumbersTab data={data} /></ErrorBoundary></TabsContent>- <TabsContent value="growth" className="mt-4">- <ErrorBoundary title="Couldn't load activity">- <GrowthTab data={data} />+ <TabsContent value="rhythm" className="mt-4">+ <ErrorBoundary title="Couldn't load rhythm">+ <RhythmTab data={data} /></ErrorBoundary></TabsContent><TabsContent value="people" className="mt-4"><ErrorBoundary title="Couldn't load contributors">- <PeopleTab data={data} />- </ErrorBoundary>- </TabsContent>- <TabsContent value="quality" className="mt-4">- <ErrorBoundary title="Couldn't load quality">- <QualityTab data={data} />- </ErrorBoundary>- </TabsContent>- <TabsContent value="architecture" className="mt-4">- <ErrorBoundary title="Couldn't load architecture">- <ArchitectureTab data={data} />+ <PeopleTab data={data} repoId={repoId} /></ErrorBoundary></TabsContent></Tabs>diff --git a/tests/unit/ingestion/test_agent_provenance.py b/tests/unit/ingestion/test_agent_provenance.pyindex 82e2729cb..22f073b35 100644--- a/tests/unit/ingestion/test_agent_provenance.py+++ b/tests/unit/ingestion/test_agent_provenance.py@@ -10,6 +10,7 @@ from __future__ import annotationsimport jsonimport time+from datetime import UTC, datetimefrom unittest.mock import MagicMockfrom repowise.core.ingestion.git_commit_index import (@@ -354,7 +355,11 @@ def _log_record(sha, an, ae, subj, body="", files=((1, 0, "src/a.py"),), ts=None, parents="") -> str:ts = ts or int(time.time())- lines = [f"\x00{sha}\x1f{an}\x1f{ae}\x1f{an}\x1f{ae}\x1f{ts}\x1f{parents}\x1f{subj}\x1f{body}"]+ iso = datetime.fromtimestamp(int(ts), UTC).isoformat()+ lines = [+ f"\x00{sha}\x1f{an}\x1f{ae}\x1f{an}\x1f{ae}\x1f{ts}\x1f{iso}"+ f"\x1f{parents}\x1f{subj}\x1f{body}"+ ]for a, d, p in files:lines.append(f"{a}\t{d}\t{p}")return "\n".join(lines)diff --git a/tests/unit/server/test_stats_bot_filter.py b/tests/unit/server/test_stats_bot_filter.pynew file mode 100644index 000000000..ef92e0925--- /dev/null+++ b/tests/unit/server/test_stats_bot_filter.py@@ -0,0 +1,63 @@+"""Automation is excluded from the Stats page's people sections.++A bot's "peak hour" is a cron schedule and its "arrival" is the day someone+enabled it — neither is a fact about how anyone works, and leaving dependabot in+a night-owl leaderboard makes the whole list read as noise.++The precision direction matters more than recall here: a missed bot is one odd+row, but a human misread as automation disappears from the page entirely. Hence+the human cases below, all of which contain an agent's name as a substring.+"""++from __future__ import annotations++import pytest++from repowise.server.routers.stats import _is_bot++AGENTS = [+ # Coding agents, recognised via the ingestion layer's provenance registry.+ ("Claude", "[email protected]"),+ ("claude[bot]", "claude[bot]@users.noreply.github.com"),+ ("claude-code", "[email protected]"),+ ("Cursor Agent", "[email protected]"),+ ("cursor", "[email protected]"),+ ("codex", "[email protected]"),+ ("Codex", "[email protected]"),+ ("copilot-swe-agent[bot]", "copilot-swe-agent[bot]@users.noreply.github.com"),+ ("devin-ai-integration[bot]", "devin-ai-integration[bot]@users.noreply.github.com"),+ ("gemini", "[email protected]"),+ # CI automation, which provenance has no reason to model.+ ("dependabot[bot]", "49699333+dependabot[bot]@users.noreply.github.com"),+ ("github-actions[bot]", "[email protected]"),+ ("renovate", "[email protected]"),+]++HUMANS = [+ ("Raghav Chamadiya", "[email protected]"),+ # Each of these contains an agent token as a substring.+ ("Claudia Silva", "[email protected]"),+ ("Jean-Claude Dupont", "[email protected]"),+ ("Aiden Cursorly", "[email protected]"),+ ("Devin Townsend", "[email protected]"),+ ("Abbot Smith", "[email protected]"),+ ("Gemma Wright", "[email protected]"),+ # A real employee at a vendor domain is not that vendor's agent identity.+ ("Jane Doe", "[email protected]"),+]++[email protected](("name", "email"), AGENTS)+def test_automation_is_filtered(name: str, email: str) -> None:+ assert _is_bot(name, email) is True++[email protected](("name", "email"), HUMANS)+def test_humans_survive(name: str, email: str) -> None:+ assert _is_bot(name, email) is False+++def test_blank_identity_is_not_automation() -> None:+ """An unattributed commit is a miss, not a bot — never drop it silently."""+ assert _is_bot(None, None) is False+ assert _is_bot("", "") is Falsediff --git a/tests/unit/server/test_stats_contributors.py b/tests/unit/server/test_stats_contributors.pyindex 618e51011..16e1dc7f5 100644--- a/tests/unit/server/test_stats_contributors.py+++ b/tests/unit/server/test_stats_contributors.py@@ -14,7 +14,7 @@ from httpx import AsyncClientfrom repowise.core.persistence.crud import get_repository, upsert_git_commits_bulkfrom repowise.core.persistence.database import get_session-from repowise.server.routers.stats import _activity+from repowise.server.routers.stats import _commit_passfrom tests.unit.server.conftest import create_test_repo@@ -53,7 +53,7 @@ async def test_contributor_count_folds_noreply_and_same_name(client: AsyncClientasync with get_session(app.state.session_factory) as session:repo_row = await get_repository(session, repo["id"])- activity = await _activity(session, repo["id"], repo_row)+ activity = (await _commit_pass(session, repo["id"], repo_row))["origin"]# No whole-history totals stored on the repo, so the count folds the# bounded sample's author identities (Jane + Bob, not 4).@@ -85,7 +85,7 @@ async def test_activity_prefers_whole_history_totals(client: AsyncClient, app) -async with get_session(app.state.session_factory) as session:repo_row = await get_repository(session, repo["id"])- activity = await _activity(session, repo["id"], repo_row)+ activity = (await _commit_pass(session, repo["id"], repo_row))["origin"]assert activity["total_commits"] == 5000assert activity["contributor_count"] == 42diff --git a/tests/unit/server/test_stats_signals.py b/tests/unit/server/test_stats_signals.pyindex 61ed70e5a..f2ad5d8a8 100644--- a/tests/unit/server/test_stats_signals.py+++ b/tests/unit/server/test_stats_signals.py@@ -1,9 +1,8 @@-"""New "By the Numbers" signals: punch card, commit velocity, change-risk mix,-truck factor, dependency cycles, and the wiki build meta.+"""Unit tests for the Stats page's derived signals.-The pure summarizers are unit-tested directly; the commit-scan signals go-through ``_activity`` on a seeded ``git_commits`` table so the wiring (SELECT-columns, bucketing) is covered end to end.+Covers the punch-card fold, commit velocity, the truck-factor rollup, and the+single ``_commit_pass`` scan that produces the origin / rhythm / people sections+from one read of the commits table."""from __future__ import annotations@@ -17,7 +16,7 @@ from httpx import AsyncClientfrom repowise.core.persistence.crud import get_repository, upsert_git_commits_bulkfrom repowise.core.persistence.database import get_sessionfrom repowise.server.routers.stats import (- _activity,+ _commit_pass,_commit_velocity,_people,_punch_card_summary,@@ -57,7 +56,7 @@ def test_punch_card_summary_peak() -> None:punch[6][12] = 2 # Sunday (weekend)dated = 10 + 4 + 3 + 2- out = _punch_card_summary(punch, dated_total=dated)+ out = _punch_card_summary(punch, dated_total=dated, timezone_mode="utc")assert out["peak"] == {"weekday": 2, "hour": 14, "count": 10}assert out["busiest_weekday"] == 2@@ -66,7 +65,7 @@ def test_punch_card_summary_peak() -> None:def test_punch_card_summary_empty() -> None:- out = _punch_card_summary([[0] * 24 for _ in range(7)], dated_total=0)+ out = _punch_card_summary([[0] * 24 for _ in range(7)], dated_total=0, timezone_mode="utc")assert out["peak"] is Noneassert out["busiest_weekday"] is Noneassert out["total"] == 0@@ -102,19 +101,17 @@ def _meta(owner: str, path: str) -> SimpleNamespace:return SimpleNamespace(primary_owner_name=owner, bus_factor=1, file_path=path)-async def test_truck_factor_single_dominant_owner() -> None:+def test_truck_factor_single_dominant_owner() -> None:# One owner holds 8/10 files → truck factor 1.metas = [_meta("Ada", f"a{i}.py") for i in range(8)] + [_meta("Bob", "b.py"),_meta("Cara", "c.py"),]- out = await _people(None, "repo", metas)+ out = _people(metas)assert out["truck_factor"] == 1-async def test_truck_factor_spread_ownership() -> None:+def test_truck_factor_spread_ownership() -> None:# Four owners with 3/3/2/2 → need 2 to cross 50% of 10.metas = ([_meta("Ada", f"a{i}.py") for i in range(3)]@@ -122,23 +119,22 @@ async def test_truck_factor_spread_ownership() -> None:+ [_meta("Cara", f"c{i}.py") for i in range(2)]+ [_meta("Dan", f"d{i}.py") for i in range(2)])- out = await _people(None, "repo", metas)+ out = _people(metas)assert out["truck_factor"] == 2-async def test_truck_factor_none_without_owners() -> None:- out = await _people(None, "repo", [SimpleNamespace(primary_owner_name=None, bus_factor=0, file_path="x.py")])+def test_truck_factor_none_without_owners() -> None:+ out = _people([SimpleNamespace(primary_owner_name=None, bus_factor=0, file_path="x.py")])assert out["truck_factor"] is None# ----------------------------------------------------------------------------# End-to-end through _activity on a seeded commits table+# End-to-end through _commit_pass on a seeded commits table# ---------------------------------------------------------------------------@pytest.mark.asyncio-async def test_activity_punch_card_and_risk_mix(client: AsyncClient, app) -> None:+async def test_punch_card_buckets_commits(client: AsyncClient, app) -> None:repo = await create_test_repo(client)rows = [# Wednesday 2024-01-03 at 14:00 UTC — weekday 2, hour 14.@@ -154,38 +150,65 @@ async def test_activity_punch_card_and_risk_mix(client: AsyncClient, app) -> Nonasync with get_session(app.state.session_factory) as session:repo_row = await get_repository(session, repo["id"])- activity = await _activity(session, repo["id"], repo_row)+ out = await _commit_pass(session, repo["id"], repo_row)- pc = activity["punch_card"]+ pc = out["rhythm"]["punch_card"]assert pc["matrix"][2][14] == 2 # both Wednesday-2pm commits landed in one cellassert pc["peak"] == {"weekday": 2, "hour": 14, "count": 2}# Weekend rows carry their commits; which days count as the weekend is a# reader preference resolved in the UI, not here.assert pc["matrix"][5][11] + pc["matrix"][6][12] == 2-- assert activity["change_risk_mix"] == {"low": 1, "moderate": 1, "high": 2}+ # No commit carries an offset, so the card stays honestly in UTC rather+ # than mixing clocks.+ assert pc["timezone_mode"] == "utc"@pytest.mark.asyncio-async def test_graph_and_build_degrade_to_zero(client: AsyncClient, app) -> None:- """The graph + build sections return a fully-shaped zero payload on a repo- with no graph-membership or LLM-cost rows (index-only / brand-new repo)."""- from repowise.server.routers.stats import _build, _graph+async def test_commit_pass_degrades_on_empty_repo(client: AsyncClient, app) -> None:+ """Every section is fully shaped on a repo with no commits at all, so the+ page renders an empty state rather than throwing on a missing key."""+ repo = await create_test_repo(client)+ async with get_session(app.state.session_factory) as session:+ repo_row = await get_repository(session, repo["id"])+ out = await _commit_pass(session, repo["id"], repo_row)++ assert out["rhythm"]["punch_card"]["total"] == 0+ assert out["rhythm"]["punch_card"]["peak"] is None+ assert out["rhythm"]["longest_streak"] is None+ assert out["rhythm"]["busiest_day"] is None+ assert out["rhythm"]["velocity"] == {"recent_90d": 0, "prior_90d": 0, "pct_change": None}+ assert out["chronotypes"] == []+ assert out["arrivals"] == []+ assert out["origin"]["age_days"] is None++async def test_punch_card_uses_author_local_time_when_offsets_present(+ client: AsyncClient, app+) -> None:+ """With every commit carrying its UTC offset, hours are shifted into the+ author's own clock — the whole point of storing the offset. A commit at+ 18:30 UTC with a +05:30 offset is midnight local, i.e. the next weekday."""repo = await create_test_repo(client)+ rows = [+ {+ **_commit("a1", datetime(2024, 1, 3, 18, 30, tzinfo=UTC)),+ "committed_offset_minutes": 330,+ },+ {+ **_commit("a2", datetime(2024, 1, 3, 18, 45, tzinfo=UTC)),+ "committed_offset_minutes": 330,+ },+ ]async with get_session(app.state.session_factory) as session:- graph = await _graph(session, repo["id"])- build = await _build(session, repo["id"])-- assert graph == {- "cycle_clusters": 0,- "files_in_cycles": 0,- "largest_cycle": 0,- "community_count": 0,- }- assert build == {- "page_count": 0,- "total_tokens": 0,- "cost_usd": 0.0,- "llm_operations": 0,- }+ await upsert_git_commits_bulk(session, repo["id"], rows)++ async with get_session(app.state.session_factory) as session:+ repo_row = await get_repository(session, repo["id"])+ out = await _commit_pass(session, repo["id"], repo_row)++ pc = out["rhythm"]["punch_card"]+ assert pc["timezone_mode"] == "author_local"+ # Wednesday 18:30 UTC + 5:30 = Thursday 00:00 local → weekday 3, hour 0.+ assert pc["matrix"][3][0] == 2+ assert pc["matrix"][2][18] == 0diff --git a/tests/unit/server/test_stats_superlatives.py b/tests/unit/server/test_stats_superlatives.pyindex 4ce3afaeb..99ba17512 100644--- a/tests/unit/server/test_stats_superlatives.py+++ b/tests/unit/server/test_stats_superlatives.py@@ -14,7 +14,7 @@ from httpx import AsyncClientfrom repowise.core.persistence.crud import get_repository, upsert_git_commits_bulkfrom repowise.core.persistence.database import get_session-from repowise.server.routers.stats import _activity+from repowise.server.routers.stats import _commit_passfrom tests.unit.server.conftest import create_test_repo_DAY = 86400@@ -59,7 +59,7 @@ async def test_biggest_commit_skips_initial_and_streak_counts_days(async with get_session(app.state.session_factory) as session:repo_row = await get_repository(session, repo["id"])- activity = await _activity(session, repo["id"], repo_row)+ activity = await _commit_pass(session, repo["id"], repo_row)biggest = activity["biggest_commit"]assert biggest is not None@@ -67,6 +67,6 @@ async def test_biggest_commit_skips_initial_and_streak_counts_days(assert biggest["lines_changed"] == 1000assert biggest["files_changed"] == 12- streak = activity["longest_streak"]+ streak = activity["rhythm"]["longest_streak"]assert streak is not Noneassert streak["days"] == 4diff --git a/tests/unit/test_git_commit_index.py b/tests/unit/test_git_commit_index.pyindex c5bc6f592..68c1f6afc 100644--- a/tests/unit/test_git_commit_index.py+++ b/tests/unit/test_git_commit_index.py@@ -3,11 +3,17 @@from __future__ import annotationsimport time+from datetime import UTC, datetimefrom unittest.mock import MagicMockfrom repowise.core.ingestion.git_commit_index import load_commit_index+def _iso_from_ts(ts: int) -> str:+ """Render a git ``%cI`` value (strict ISO-8601 with a numeric offset)."""+ return datetime.fromtimestamp(int(ts), UTC).isoformat()++def _build_log(commits: list[dict]) -> str:"""Render an in-memory list of commit dicts into the raw `git log` format.@@ -29,6 +35,8 @@ def _build_log(commits: list[dict]) -> str:+ "\x1f"+ str(c["ct"])+ "\x1f"+ + _iso_from_ts(c["ct"])+ + "\x1f"+ c.get("parents", "")+ "\x1f"+ c["subj"]diff --git a/tests/unit/test_git_commit_rows.py b/tests/unit/test_git_commit_rows.pyindex 9f1f0a535..6754b31ce 100644--- a/tests/unit/test_git_commit_rows.py+++ b/tests/unit/test_git_commit_rows.py@@ -8,12 +8,18 @@ just-in-time change-risk, ordering, truncation).from __future__ import annotationsimport time+from datetime import UTC, datetimefrom unittest.mock import MagicMockfrom repowise.core.ingestion.git_commit_index import load_commit_indexfrom repowise.core.ingestion.git_indexer.commit_rows import build_commit_rows+def _iso_from_ts(ts: int) -> str:+ """Render a git ``%cI`` value (strict ISO-8601 with a numeric offset)."""+ return datetime.fromtimestamp(int(ts), UTC).isoformat()++def _build_log(commits: list[dict]) -> str:lines: list[str] = []for c in commits:@@ -31,6 +37,8 @@ def _build_log(commits: list[dict]) -> str:+ "\x1f"+ str(c["ct"])+ "\x1f"+ + _iso_from_ts(c["ct"])+ + "\x1f"+ c.get("parents", "")+ "\x1f"+ c["subj"]@@ -87,12 +95,30 @@ def test_commit_sink_since_ts_drops_old_commits() -> None:the incremental capture path's freshness guarantee."""raw = _build_log([- {"sha": "new", "an": "A", "ae": "a@x", "ct": 3000, "subj": "new",- "files": [(1, 0, "src/a.py")]},- {"sha": "mid", "an": "A", "ae": "a@x", "ct": 2000, "subj": "mid",- "files": [(1, 0, "src/a.py")]},- {"sha": "old", "an": "A", "ae": "a@x", "ct": 1000, "subj": "old",- "files": [(1, 0, "src/a.py")]},+ {+ "sha": "new",+ "an": "A",+ "ae": "a@x",+ "ct": 3000,+ "subj": "new",+ "files": [(1, 0, "src/a.py")],+ },+ {+ "sha": "mid",+ "an": "A",+ "ae": "a@x",+ "ct": 2000,+ "subj": "mid",+ "files": [(1, 0, "src/a.py")],+ },+ {+ "sha": "old",+ "an": "A",+ "ae": "a@x",+ "ct": 1000,+ "subj": "old",+ "files": [(1, 0, "src/a.py")],+ },])repo = MagicMock()diff --git a/tests/unit/test_git_indexer.py b/tests/unit/test_git_indexer.pyindex 3df6b1b34..8a7c92ccd 100644--- a/tests/unit/test_git_indexer.py+++ b/tests/unit/test_git_indexer.py@@ -46,6 +46,11 @@ def _make_diff(b_path: str, a_path: str | None = None):# ---------------------------------------------------------------------------+def _iso(ts: int) -> str:+ """Git ``%cI`` for a unix timestamp — the offset field the parser expects."""+ return datetime.fromtimestamp(int(ts), UTC).isoformat()++class TestSignificantCommitFilter:"""Merges, bumps, chore, and bot commits are filtered out; normal commits pass.@@ -391,7 +396,7 @@ class TestStableClassification:for i in range(15):ts = int((old_date - timedelta(days=i)).timestamp())log_lines.append(- f"\x00sha{i:04d}\x1fAlice\[email protected]\x1fAlice\[email protected]\x1f{ts}\x1f\x1ffeat: old commit {i}\x1f"+ f"\x00sha{i:04d}\x1fAlice\[email protected]\x1fAlice\[email protected]\x1f{ts}\x1f{_iso(ts)}\x1f\x1ffeat: old commit {i}\x1f")mock_repo.git.log.return_value = "\n".join(log_lines)@@ -420,7 +425,7 @@ class TestStableClassification:for i, (name, email, when) in enumerate(specs):ts = int(when.timestamp())log_lines.append(- f"\x00sha{i:04d}\x1f{name}\x1f{email}\x1f{name}\x1f{email}\x1f{ts}\x1f\x1ffeat: c{i}\x1f"+ f"\x00sha{i:04d}\x1f{name}\x1f{email}\x1f{name}\x1f{email}\x1f{ts}\x1f{_iso(ts)}\x1f\x1ffeat: c{i}\x1f")mock_repo.git.log.return_value = "\n".join(log_lines)@@ -448,7 +453,7 @@ class TestGitWindowAnchor:for i in range(15):ts = int((old_date - timedelta(days=i)).timestamp())log_lines.append(- f"\x00sha{i:04d}\x1fAlice\[email protected]\x1fAlice\[email protected]\x1f{ts}\x1f\x1ffeat: old {i}\x1f"+ f"\x00sha{i:04d}\x1fAlice\[email protected]\x1fAlice\[email protected]\x1f{ts}\x1f{_iso(ts)}\x1f\x1ffeat: old {i}\x1f")mock_repo.git.log.return_value = "\n".join(log_lines)# HEAD tip = the newest of the (old) batch.@@ -677,7 +682,9 @@ class TestNumstatParsing:f"\x1f{e.get('email', '[email protected]')}"f"\x1f{e.get('committer', e.get('author', 'Dev'))}"f"\x1f{e.get('committer_email', e.get('email', '[email protected]'))}"- f"\x1f{e['ts']}\x1f{e.get('parents', '')}"+ f"\x1f{e['ts']}"+ f"\x1f{datetime.fromtimestamp(int(e['ts']), UTC).isoformat()}"+ f"\x1f{e.get('parents', '')}"f"\x1f{e.get('subject', 'some commit')}"f"\x1f{e.get('body', '')}")@@ -765,7 +772,7 @@ class TestNumstatParsing:# First record is malformed (only 3 fields), second is validraw = (f"\x00badsha\x1fAlice\[email protected]\n" # only 3 fields- f"\x00goodsha\x1fBob\[email protected]\x1fBob\[email protected]\x1f{recent_ts}\x1f\x1ffeat: valid commit\x1f\n"+ f"\x00goodsha\x1fBob\[email protected]\x1fBob\[email protected]\x1f{recent_ts}\x1f{_iso(recent_ts)}\x1f\x1ffeat: valid commit\x1f\n"f"10\t5\ttest.py\n")mock_repo.git.log.return_value = raw@@ -783,7 +790,7 @@ class TestNumstatParsing:recent_ts = int((datetime.now(UTC) - timedelta(days=5)).timestamp())raw = (f"\x00sha1\x1fAlice\[email protected]\x1fAlice\[email protected]\x1f{recent_ts}"- f"\x1fparent1 parent2\x1fMerge branch main\x1f\n"+ f"\x1f{_iso(recent_ts)}\x1fparent1 parent2\x1fMerge branch main\x1f\n"f"5\t2\tmerged.py\n")mock_repo.git.log.return_value = raw@@ -803,13 +810,13 @@ class TestNumstatParsing:# Simulates --follow output: recent commit uses new name,# older commit uses rename notation, oldest uses old name.raw = (- f"\x00sha1\x1fAlice\[email protected]\x1fAlice\[email protected]\x1f{recent_ts}\x1f\x1ffeat: update\x1f\n"+ f"\x00sha1\x1fAlice\[email protected]\x1fAlice\[email protected]\x1f{recent_ts}\x1f{_iso(recent_ts)}\x1f\x1ffeat: update\x1f\n"f"10\t3\tsrc/new_name.py\n"f"\n"- f"\x00sha2\x1fAlice\[email protected]\x1fAlice\[email protected]\x1f{old_ts}\x1f\x1frename file\x1f\n"+ f"\x00sha2\x1fAlice\[email protected]\x1fAlice\[email protected]\x1f{old_ts}\x1f{_iso(old_ts)}\x1f\x1frename file\x1f\n"f"0\t0\t{{src/old_name.py => src/new_name.py}}\n"f"\n"- f"\x00sha3\x1fAlice\[email protected]\x1fAlice\[email protected]\x1f{old_ts - 86400}\x1f\x1ffeat: old work\x1f\n"+ f"\x00sha3\x1fAlice\[email protected]\x1fAlice\[email protected]\x1f{old_ts - 86400}\x1f{_iso(old_ts - 86400)}\x1f\x1ffeat: old work\x1f\n"f"20\t5\tsrc/old_name.py\n")mock_repo.git.log.return_value = raw