graff — a fast agentic coding harness in Zig: multi-provider, MCP, workflows, DGM evolution loop, TS/Python SDKs
Python
{
"id": "linear-issues",
"suite": "mcp",
"category": "mcp-shapes",
"prompt": "This workspace has a Linear-shaped MCP server named linear (stdio, deferred). Load its schemas, then fetch all 8 issues and every issue's comments. Write report.json as {\"issue_count\": N, \"issues\": [{\"id\", \"title\", \"comment_count\", \"latest_author\"}, ...]} in ISS-1..ISS-8 order. latest_author is the author.name of the newest comment (last in that issue's list). Do not invent ids or authors. If rlm is available, prefer one script after load_tool_schemas: issues = list_issues(); comments = each(issues, list_comments, \"id\").",
"files": {
".mcp.json": "{\"mcpServers\":{\"linear\":{\"command\":\"python3\",\"args\":[\"linear_fixture_mcp.py\"]}}}\n"
},
"setup": ["python3 -c \"import pathlib,shutil; p=pathlib.Path('.').resolve(); root=next(x for x in p.parents if (x/'scripts'/'linear_fixture_mcp.py').exists()); shutil.copy(root/'scripts'/'linear_fixture_mcp.py','.')\""],
"check": "python3 \"$TASK_ROOT/hidden/check_linear_report.py\"",
"timeout_s": 240
}