A self-organizing Obsidian vault that gives AI coding agents persistent memory.
TypeScript ai-agentsclaude-codecodex-cligemini-cliknowledge-managementobsidianobsidian-vaultpersistent-memorysecond-braintemplate
v2: brain/, bases, org, slash commands, hooks, Home.md
Brenno Ferraric306e323279c
diff --git a/.claude/commands/capture-1on1.md b/.claude/commands/capture-1on1.mdnew file mode 100644index 0000000..5033950--- /dev/null+++ b/.claude/commands/capture-1on1.md@@ -0,0 +1,52 @@+# Capture 1:1 Meeting++Take a meeting transcript, notes, or Gemini summary and create a structured vault note with key takeaways, quotes, action items, and DM context.++## Usage++```+/capture-1on1 <participant>+```++User will paste the transcript/notes. Example: `/capture-1on1 <name>`++## Workflow++1. **Parse the input** — handle Gemini transcripts, raw notes, or meeting summaries.++2. **Create work note** at `work/1-1/<Participant> <YYYY-MM-DD>.md` with:++ ```yaml+ ---+ date: "<meeting date>"+ description: "<one-line summary of key topics discussed>"+ tags:+ - work-note+ status: completed+ ---+ ```++3. **Structure the note** with these sections:+ - **Key Takeaways** — bullet points of the most important things discussed, grouped by topic+ - **Decisions Made** — anything agreed upon+ - **Action Items** — with checkboxes, who owns each+ - **Quotes Worth Noting** — direct quotes that reveal priorities, feedback, or dynamics (use blockquotes)+ - **What Went Well** — what landed, what resonated+ - **What to Watch** — things to monitor, concerns, ambiguous signals+ - **Context From DMs** — check relevant DMs before/after the meeting for color (ask user for channel IDs if needed)+ - **Related** — wikilinks to relevant notes++4. **Update related notes**:+ - Person's people note (add 1:1 section or update existing)+ - `work/Index.md` (add to Review Prep or Active Projects as appropriate)+ - `brain/Memories.md` if any context changed (manager dynamics, priorities, etc.)+ - `org/people/` note for the participant++5. **Check for stale context** — if the meeting reveals something that contradicts existing vault notes, flag and update.++## Important++- Preserve the conversational tone — don't over-formalize quotes+- Flag sensitive interpersonal items — note if they should stay private+- Separate what was SAID from what it MEANS (interpretation goes in "What to Watch")+- Cross-reference with DMs from the same day for fuller picturediff --git a/.claude/commands/incident-capture.md b/.claude/commands/incident-capture.mdnew file mode 100644index 0000000..fc5f3a5--- /dev/null+++ b/.claude/commands/incident-capture.md@@ -0,0 +1,111 @@+# Incident Capture++Capture an incident from Slack channels, DMs, and threads into structured vault notes. Produces a complete incident work note with timeline, people, analysis, and brag doc entry.++## Usage++```+/incident-capture <slack-urls>+```++Provide one or more Slack URLs: incident channels, DM conversations, threads. The more context, the better.++## Workflow++### 1. Gather Raw Data++For each Slack URL provided:+- Read the full channel/DM/thread using `slack_read_channel` or `slack_read_thread`+- Read ALL sub-threads within the channel (check for "Thread: N replies" indicators)+- Note every timestamp, person, and message — nothing is too small+- Download and examine any shared images or files++### 2. Identify People++For every person who posted or was mentioned:+- Fetch their Slack profile (`slack_read_user_profile`)+- Check if they have a person note in `org/people/`+- Note their role, team, and title+- Track who did what (reported, investigated, fixed, confirmed, etc.)++### 3. Build the Timeline++Reconstruct a detailed timeline from all sources:+- Every message with exact timestamp (local timezone)+- Attribution: who said/did what+- Cross-reference between channels (same person posting in incident channel + DMs)+- Key moments: first report, incident declared, root cause identified, fix created, fix merged, resolution confirmed++### 4. Create the Work Note++Create `work/incidents/<Incident Name>.md` with:++```yaml+---+date: "YYYY-MM-DD"+quarter: QN-YYYY+description: "~150 chars"+project: <relevant project>+status: active+irp: XXXX+severity: high/medium/low+role: <your role>+tags:+ - work-note+ - incident+---+```++Sections:+- **Context** — what happened, what triggered it, who was involved+- **Root Cause** — technical explanation+- **Resolution** — fix PR(s), what changed+- **Timeline** — full detailed table with timestamps+- **Impact** — users affected, business impact+- **Involved Personnel** — with wikilinks to person notes+- **Notes** — key actions by you, analysis+- **Analysis** — what this means strategically (pattern, visibility, competencies)+- **Related** — wikilinks to all related notes, competencies++### 5. Create/Update People Notes++For key people involved who don't have person notes:+- Create in `org/people/` with proper frontmatter (`date`, `title`, `description`, `team`, `tags: [person]`)+- Add to `org/People & Context.md`++For existing people notes:+- Add the incident as a Key Moment++### 6. Update Indexes++- `work/Index.md` — add to Incidents section+- `brain/Memories.md` — add incident summary to Recent Context+- `brain/Patterns.md` — if this reveals a recurring pattern+- `brain/Gotchas.md` — if this reveals a technical gotcha+- `perf/Brag Doc.md` — add to relevant quarter with competency links+- `perf/brag/QN YYYY.md` — add detailed brag entry++### 7. Prepare Post-Mortem Draft (if applicable)++If you are or may become the post-mortem manager:+- Check your org's post-mortem template+- Create `work/incidents/<Ticket> Post-Mortem Draft.md` following your org's post-mortem template+- Include: Executive Summary, Narrative Timeline (4-column), 5 Whys analysis, Action Items table++### 8. Offer Next Steps++After capturing, suggest:+- "Want me to prepare the incident tracking system fields?" (Points of Time, Impact, RCA)+- "Want me to draft a message for the incident channel?"+- "Want me to create a root cause analysis document?"+- "Should I run `/vault-audit` to verify everything links properly?"++## Important++- **Read every message** — don't skim or summarize prematurely+- **Preserve exact timestamps** — incident timelines need precision+- **Attribute everything** — who said what, who did what+- **Cross-reference** — the same event may be described differently in DMs vs incident channel+- **Check Slack profiles** — roles and teams matter for the story+- **Be blameless in public docs** — use commit SHAs, not names, in shareable documents+- **Private analysis is honest** — the vault work note can include strategic analysis about credit, visibility, etc.diff --git a/.claude/commands/peer-scan.md b/.claude/commands/peer-scan.mdnew file mode 100644index 0000000..72a17d5--- /dev/null+++ b/.claude/commands/peer-scan.md@@ -0,0 +1,53 @@+# Peer PR Deep Scan++Deep scan a peer's GitHub PRs for performance review preparation. Produces a structured analysis saved to `perf/evidence/`.++## Usage++```+/peer-scan <name> <github-username> <repo> [period]+```++Example: `/peer-scan "Jane Doe" jdoe example-repo "Oct 2025 - Mar 2026"`++## Workflow++1. **Fetch full PR list** (limit 200):+ ```+ gh pr list --repo <org>/<repo> --author <username> --state all --limit 200 --json number,title,state,createdAt,mergedAt,additions,deletions+ ```++2. **Filter to review period** (default: last 6 months). For EACH PR in period, fetch details:+ ```+ gh pr view <number> --repo <org>/<repo> --json body,reviews,comments,additions,deletions,changedFiles,title,state,createdAt,mergedAt+ ```++3. **Produce structured analysis** with these sections:+ - **PR count by month** — table showing velocity trends+ - **Projects/themes** — group PRs by project area with descriptions of what was built+ - **Quality signals** — review comments, change requests, approval patterns, reverts, SonarQube+ - **Notable contributions** — architectural decisions, complex fixes, test coverage, cross-team impact+ - **Growth signals** — scope expansion over time, leadership evidence+ - **Full PR table** — every PR with number, title, date, additions, deletions, state++4. **Save to vault** as `perf/evidence/<Name> PRs - <Period>.md` with frontmatter:+ ```yaml+ ---+ date: "<today>"+ description: "<one-line summary of findings>"+ person: "<Full Name>"+ cycle: "<e.g. h1-2026>"+ tags:+ - perf+ - evidence+ ---+ ```++## Important++- Be thorough — this feeds into performance reviews+- Note PRs that were reverted or closed (quality signal)+- Look for patterns in reviewer feedback (recurring issues = growth area)+- Identify cross-team PRs (collaboration evidence)+- Flag weekend/late-night work patterns if visible+- Map PRs to projects you have context ondiff --git a/.claude/commands/project-archive.md b/.claude/commands/project-archive.mdnew file mode 100644index 0000000..9bcb7bd--- /dev/null+++ b/.claude/commands/project-archive.md@@ -0,0 +1,47 @@+# Project Archive++Move a completed project from `work/active/` to `work/archive/YYYY/` and update all indexes.++## Usage++```+/project-archive <project name>+```++## Workflow++### 1. Find the Note++Search `work/active/` for the project name. Confirm with the user before proceeding.++### 2. Update Frontmatter++- Set `status: completed`+- Verify `quarter` property is set correctly+- Verify `description` reflects the final state++### 3. Move the File++```bash+git mv "work/active/<Note>.md" "work/archive/YYYY/"+```++Use the year from the note's `date` field.++### 4. Update Indexes++- **`work/Index.md`**: Move from Active Projects to the appropriate Completed quarter section+- **`brain/North Star.md`**: Mark as completed in Current Focus if listed there+- **`perf/Brag Doc.md`**: Verify the project is captured in the relevant quarter's highlights+- **`brain/Memories.md`**: Update Recent Context if the project is mentioned as "in progress"++### 5. Verify++- Run a quick check that no wikilinks are broken (Obsidian resolves by name, so moves shouldn't break links)+- Confirm the Work Dashboard Base shows the note in "Completed" view, not "Active Work"++## Important++- Always use `git mv` — never copy+delete+- Don't archive without user confirmation+- If the project has sub-notes (like incidents have RCA + deep dive), ask if those should move toodiff --git a/.claude/commands/review-brief.md b/.claude/commands/review-brief.mdnew file mode 100644index 0000000..f4fbd94--- /dev/null+++ b/.claude/commands/review-brief.md@@ -0,0 +1,63 @@+# Generate Review Brief++Generate a performance review context transfer document from vault data. Supports manager version (manager-friendly) and peer version (project-focused).++## Usage++```+/review-brief <audience> [period]+```++Examples:+- `/review-brief manager "Q4 2025 + Q1 2026"`+- `/review-brief peers "Q4 2025 + Q1 2026"`++## Workflow++### 1. Gather Data++Read these vault sources:+- `perf/<cycle>/<cycle> Review Brief.md` (or current private brief) — full context+- `perf/Brag Doc.md` — quarterly highlights+- `perf/brag/Q*.md` — quarterly detail notes for the period+- `perf/evidence/<Your Name> PRs - *.md` — PR data+- `work/*.md` — project notes for the period+- `perf/competencies/*.md` — competency definitions+- Previous review notes for baseline comparison++### 2. Generate Content++**For manager audience:**+- Frame for a non-technical audience — outcome language, not technical jargon+- Include: The Arc (narrative), Impact at a Glance (table), Impact Details (per project), Competency Highlights (with baselines), Documentation Trail+- Replace technical terms: "deadlock" → "timing conflict", "data race" → "concurrency issue", etc.+- No wikilinks — use plain text or markdown links to external resources+- Include all documentation, issue tracker, source control, monitoring, and messaging references++**For peer audience:**+- Can be more technical but still manager-friendly (peers write reviews that go to your manager)+- Organize by project (matches your org's review tool structure)+- Include "Other things worth mentioning" for non-project work+- Casual tone — "jog your memory", "no pressure to cover everything"+- No competency section — that's for your manager++### 3. Create Files++- Markdown version in `perf/`+- HTML version with professional styling (blue theme, tables, responsive)+- PDF via Chrome headless: `--headless --no-pdf-header-footer --print-to-pdf`++### 4. Verify++- Check page breaks in PDF (render pages with pdftoppm)+- Ensure no content is cut mid-section+- Verify all links work+- Cross-check PR counts and dates against `reference/` data++## Important++- NEVER include: sensitive interpersonal details, private 1:1 talking points, peer selection strategy, personal strategic notes in shared versions+- Always maintain a private version with full context+- When updating, update BOTH private + shared versions+- Manager version: no wikilinks, non-technical language, professional formatting+- Peer version: project-focused, same manager-friendly languagediff --git a/.claude/commands/slack-scan.md b/.claude/commands/slack-scan.mdnew file mode 100644index 0000000..d9e752c--- /dev/null+++ b/.claude/commands/slack-scan.md@@ -0,0 +1,51 @@+# Slack Deep Scan++Deep scan Slack channels and DMs for evidence involving a specific person or project. Extracts every touchpoint with timestamps for review evidence or project documentation.++## Usage++```+/slack-scan <target> [channels...] [date-range]+```++Examples:+- `/slack-scan "Jane Doe" C0EXAMPLE1 C0EXAMPLE2 "after:2026-03-16"`+- `/slack-scan "project:example" C0EXAMPLE1 C0EXAMPLE2 C0EXAMPLE3 C0EXAMPLE4`++## Workflow++1. **Read each channel** fully (paginate with cursor until all messages are fetched):+ ```+ mcp slack_read_channel channel_id=<id> limit=100+ ```++2. **Search for the person** across public+private channels:+ ```+ mcp slack_search_public_and_private query="from:<@USER_ID> after:<date>"+ mcp slack_search_public_and_private query="<Name> after:<date> -from:<@USER_ID>"+ ```++3. **Check DMs** the user points to. Read full history for the relevant period.++4. **For each message involving the target**, extract:+ - Timestamp and channel+ - What was said/done+ - Context (what prompted it, what followed)+ - Whether it's evidence of: technical work, leadership, collaboration, problem-solving, initiative++5. **Organize by date** with clear headers per day. Separate:+ - **Project evidence** → goes to the relevant work note+ - **Review/personal context** → goes to people notes or review prep notes+ - **Team dynamics** → goes to people notes++6. **Flag items that need correction** in existing vault notes.++## Important++- Be meticulous — timestamps matter for evidence+- Capture exact quotes when they show initiative, leadership, or problem-solving+- Note when the person was tagged by others (shows they're a go-to person)+- Separate project work from review prep from personal conversations+- Don't mix contexts — put data where it belongs in the vault+- Check for threads (slack_read_thread) when a message has replies+- Look for screen recordings, files shared, PRs linkeddiff --git a/.claude/commands/vault-audit.md b/.claude/commands/vault-audit.mdnew file mode 100644index 0000000..960c8c0--- /dev/null+++ b/.claude/commands/vault-audit.md@@ -0,0 +1,133 @@+# Vault Audit++Deep structural audit of the vault. Checks indexes, folder placement, frontmatter, links, Bases, and consistency. Fix what can be fixed, flag what needs user input.++**When to use**: After substantial sessions, after reorganization, or periodically to maintain vault health. For lighter end-of-session checks, use `/wrap-up` instead.++## Usage++```+/vault-audit+```++## Workflow++### 1. Check Folder Structure++Verify the vault matches the expected layout:+- `Home.md` exists at vault root+- `bases/` contains all `.base` files (none scattered elsewhere)+- `work/active/` contains only notes with `status: active`+- `work/archive/2025/` and `work/archive/2026/` contain only `status: completed` notes+- `work/incidents/` contains only notes tagged `incident`+- `work/1-1/` contains only 1:1 meeting notes+- `org/people/` contains only notes tagged `person`+- `org/teams/` contains only notes tagged `team`+- `templates/` contains only template files (with `{{placeholders}}`)+- `thinking/` is clean (no leftover drafts that should have been promoted)+- Nothing unexpected at vault root (only `Home.md`, `CLAUDE.md`, `.gitignore`)++### 2. Check Indexes++Read and verify each index file:+- `Home.md` — do embedded Base views reference existing Bases? Are quick links valid?+- `work/Index.md` — are active projects still active? Are completed items in the right section? Any missing notes?+- `brain/Memories.md` — is the "Recent Context" section current? Any stale claims?+- `org/People & Context.md` — are roles, peer selections, and project assignments current?+- `perf/Brag Doc.md` — do PR counts and project descriptions match reality?+- `brain/Skills.md` — are all slash commands registered? Workflows still valid?++### 3. Check Frontmatter Completeness++For each note type, verify required properties:++**Work notes** (`work/active/`, `work/archive/`):+- Required: `date`, `quarter`, `description`, `status`, `tags: [work-note]`+- Optional: `project`, `team`++**Incident notes** (`work/incidents/`):+- Required: `date`, `quarter`, `description`, `tags: [work-note, incident]`+- Required for main incident notes: `irp`, `severity`, `role`, `status`++**Person notes** (`org/people/`):+- Required: `date`, `title`, `description`, `tags: [person]`+- Optional but recommended: `team`++**Team notes** (`org/teams/`):+- Required: `date`, `description`, `tags: [team]`++**Brain notes** (`brain/`):+- Required: `description`, `tags: [brain]`++**1:1 notes** (`work/1-1/`):+- Required: `date`, `quarter`, `description`, `tags: [work-note]`++### 4. Check for Duplicate Tags++Scan all notes for duplicate entries in the `tags` array (e.g., `tags: [person, person]`). This is a known issue — fix any found.++### 5. Check Status/Folder Alignment++- Notes in `work/active/` must have `status: active`+- Notes in `work/archive/` must have `status: completed`+- No `status: active` notes in archive, no `status: completed` notes in active++### 6. Check Bases++For each `.base` file in `bases/`:+- Do filters still match the expected notes?+- Are templates excluded? (filters should include `!file.inFolder("templates")` where relevant)+- Do referenced properties exist in the target notes?+- Do formula references exist?++### 7. Check for Orphans++- Are there notes in `work/active/` or `work/archive/` not linked from `work/Index.md`?+- Are there incident notes not linked from `work/Index.md` Incidents section?+- Are there people notes not linked from `org/People & Context.md`?+- Are there notes without any inbound links at all? (Use `obsidian orphans` if available, or grep for `[[NoteName]]` references)+- Are there thinking notes that should have been promoted or deleted?++### 8. Check Links++- Scan for wikilinks that reference notes that don't exist (broken links)+- Check that bidirectional links exist where expected (work note ↔ person, work note ↔ competency)+- Verify `## Related` sections aren't empty on work notes++### 9. Check for Stale Context++- Read `brain/Memories.md` "Recent Context" — is anything outdated?+- Read `org/People & Context.md` — any roles, teams, or relationships that changed?+- Check `brain/Key Decisions.md`, `brain/Patterns.md`, `brain/Gotchas.md` for outdated claims+- Check `brain/North Star.md` — does Current Focus reflect reality?++### 10. Check for Mixed Context++Per vault rules, each note should cover ONE concept. Flag notes that:+- Mix project work with review prep+- Mix personal conversations with project evidence+- Have 3+ independent sections that don't need each other++### 11. Check Claude Config++- `.claude/settings.json` — are hooks well-formed and referencing correct paths?+- `.claude/commands/` — do all commands reference correct folder structure?+- `CLAUDE.md` — any stale instructions that contradict current vault state?++### 12. Fix and Report++- Fix what's clearly wrong (broken links, missing frontmatter, duplicate tags, wrong folder)+- For ambiguous issues, list them and ask the user+- Summarize:+ - **Fixed**: issues resolved+ - **Flagged**: needs user input+ - **Suggested**: improvements for the vault++## Important++- Don't delete anything without asking+- Don't create new notes during audit — just fix existing ones+- Preserve existing frontmatter when editing+- If a note is in the wrong folder, move it with `git mv`+- Update `brain/Memories.md` index if memory topics changed+- Use parallel agents for large audits (e.g., one checking work/, one checking org/, one checking perf/)diff --git a/.claude/commands/wrap-up.md b/.claude/commands/wrap-up.mdnew file mode 100644index 0000000..f1fa762--- /dev/null+++ b/.claude/commands/wrap-up.md@@ -0,0 +1,83 @@+# Wrap Up++Full session review before ending. Review context, ways of working, files modified, consistency, and suggest improvements.++## Usage++```+/wrap-up+```++Triggered when the user says "wrap up", "let's wrap", "wrapping up", or similar. Claude should invoke this automatically.++## Workflow++### 1. Review What Was Done++Scan the conversation for:+- Notes created or modified (list them all with paths)+- People notes created or updated+- Indexes updated+- Brag doc entries added+- Brain notes updated (Patterns, Gotchas, Key Decisions, Memories)++### 2. Verify Note Quality++For each note created or modified this session:+- Frontmatter complete? (`date`, `quarter`, `description`, `tags`, type-specific fields)+- At least one wikilink to another note?+- In the correct folder? (`work/active/` vs `work/archive/` vs `work/incidents/` etc.)+- Description accurate and ~150 chars?+- Status field correct?++### 3. Check Index Consistency++- `work/Index.md` — are new notes linked? Are completed projects in the right section?+- `brain/Memories.md` — does Recent Context reflect what happened this session?+- `org/People & Context.md` — any new people or relationship changes to capture?+- `perf/Brag Doc.md` — any wins or achievements from this session?+- `Home.md` — are embedded Bases still valid?++### 4. Check for Orphans++- Any new notes not linked from at least one other note?+- Any new people not added to People & Context?+- Any thinking notes that should be promoted or deleted?++### 5. Archive Check++- Are there notes in `work/active/` that should be moved to `work/archive/YYYY/`?+- Any status fields still `active` that should be `completed`?++### 6. Ways of Working Review++Check if this session revealed:+- A new pattern that should be in `brain/Patterns.md`?+- A new gotcha that should be in `brain/Gotchas.md`?+- A workflow improvement for `brain/Skills.md`?+- A CLAUDE.md update needed (new convention, stale reference)?+- A new or improved slash command?+- A hook that should be added or modified?++### 7. Suggest Improvements++Based on how the session went:+- Were there friction points in the workflow?+- Did we do something manually that could be automated?+- Did we repeat a pattern that should be a skill?+- Are there Bases that should be created or updated?+- Any frontmatter properties that would help future queries?++### 8. Report++Present a concise summary:+- **Done**: what was captured this session+- **Fixed**: issues found and resolved+- **Flagged**: things that need user input+- **Suggested**: improvements for next time++## Important++- This is a READ + VERIFY pass, not a creation pass. Fix small issues (broken links, missing frontmatter), but flag larger changes for user approval.+- Be honest about what's missing — the goal is leaving the vault in a better state than you found it.+- If North Star goals shifted during the session, suggest updating it.diff --git a/.claude/settings.json b/.claude/settings.jsonindex d685af1..1ebcf4b 100644--- a/.claude/settings.json+++ b/.claude/settings.json@@ -8,6 +8,35 @@"type": "command","command": "find \"$CLAUDE_PROJECT_DIR\" -not -path '*/.git/*' -not -path '*/.obsidian/*' -not -path '*/.claude/*' -not -path '*/.claude-plugin/*' -not -name '.DS_Store' -not -name '.gitignore' -type f | sed \"s|$CLAUDE_PROJECT_DIR/||\" | sort","timeout": 5+ },+ {+ "type": "command",+ "command": "echo 'VAULT MODE: Obsidian Mind vault active.\n- Entry point: Home.md | Bases: bases/ | Active work: work/active/\n- Load obsidian-markdown skill before editing .md files\n- Every note needs: date, quarter, description, tags, at least one wikilink\n- New work notes → work/active/ | Completed → work/archive/YYYY/ | Incidents → work/incidents/\n- People → org/people/ | Brain → brain/ | Reference → reference/\n- Use /incident-capture for incidents, /project-archive when completing projects'",+ "timeout": 2+ }+ ]+ }+ ],+ "PreToolUse": [+ {+ "matcher": "Write",+ "hooks": [+ {+ "type": "command",+ "command": "if echo \"$TOOL_INPUT\" | grep -q '\\.md\"'; then echo 'Obsidian note check: frontmatter (date, quarter, description, tags), wikilinks to at least one other note, correct folder per CLAUDE.md. Templates folder is excluded from Bases.'; fi",+ "timeout": 2+ }+ ]+ }+ ],+ "Stop": [+ {+ "matcher": "",+ "hooks": [+ {+ "type": "command",+ "command": "echo 'Session end checklist:\n- Archive completed projects? (work/active/ → work/archive/YYYY/)\n- Update indexes? (Index.md, Memories.md, People & Context, Brag Doc)\n- New notes linked? (orphans are bugs)\n- Run /vault-audit if many notes were created/modified'",+ "timeout": 2}]}diff --git a/.claude/update-skills.sh b/.claude/update-skills.shnew file mode 100755index 0000000..5a13498--- /dev/null+++ b/.claude/update-skills.sh@@ -0,0 +1,26 @@+#!/bin/bash+# Update obsidian-skills from upstream (kepano/obsidian-skills)+# Run from vault root: .claude/update-skills.sh++set -e++REPO="kepano/obsidian-skills"+SKILLS_DIR=".claude/skills"+TEMP_DIR=$(mktemp -d)++echo "Fetching latest obsidian-skills from $REPO..."+git clone --depth 1 "https://github.com/$REPO.git" "$TEMP_DIR" 2>/dev/null++for skill in "$TEMP_DIR/skills/"*/; do+ skill_name=$(basename "$skill")+ if [ -d "$SKILLS_DIR/$skill_name" ]; then+ echo "Updating $skill_name..."+ cp -R "$skill"* "$SKILLS_DIR/$skill_name/"+ else+ echo "New skill found: $skill_name — copying..."+ cp -R "$skill" "$SKILLS_DIR/$skill_name"+ fi+done++rm -rf "$TEMP_DIR"+echo "Done. Skills updated to latest."diff --git a/CLAUDE.md b/CLAUDE.mdindex 1da1294..2f79c14 100644--- a/CLAUDE.md+++ b/CLAUDE.md@@ -12,16 +12,44 @@ This vault has [obsidian-skills](https://github.com/kepano/obsidian-skills) inst- **obsidian-bases**: Create `.base` files with views, filters, and formulas. Bases core plugin is enabled. See `references/FUNCTIONS_REFERENCE.md`.- **defuddle**: Extract clean markdown from web pages via `defuddle parse <url> --md`.+### Custom Slash Commands++Defined in `.claude/commands/`. See [[Skills]] for full documentation.++| Command | Purpose |+|---------|---------|+| `/peer-scan` | Deep scan a peer's GitHub PRs for review prep |+| `/slack-scan` | Deep scan Slack channels/DMs for evidence |+| `/capture-1on1` | Capture 1:1 meeting transcript into structured vault note |+| `/vault-audit` | Audit indexes, links, orphans, stale context |+| `/review-brief` | Generate review brief (manager or peer version) |+| `/incident-capture` | Capture incident from Slack channels/DMs into structured vault notes |+| `/project-archive` | Move completed project from active/ to archive/, update indexes |+| `/wrap-up` | Full session review -- verify notes, indexes, links, suggest improvements |+## Vault Structure| Folder | Purpose | Key Files ||--------|---------|-----------|-| `work/` | Work notes, project tracking | `Index.md` (MOC) |-| `perf/` | Performance reviews, brag doc | `Brag Doc.md`, `Review Template.md` |-| `perf/competencies/` | Atomic competency notes (link targets for work notes) | One note per competency |-| `claude/` | Claude-facing context | `Memories.md` (index), topic memory notes, `Skills.md`, `North Star.md` |-| `thinking/` | Claude's scratchpad for drafts and reasoning | Named `YYYY-MM-DD-topic.md` |-| `templates/` | Obsidian templates | `Work Note.md`, `Decision Record.md`, `Thinking Note.md`, `Competency Note.md` |+| `Home.md` | **Vault entry point** -- embedded Base views, quick links | Open this first |+| `bases/` | **All Bases centralized** -- dynamic views for navigation | `Work Dashboard`, `Incidents`, `People Directory`, `1-1 History`, `Review Evidence`, `Competency Map`, `Templates` |+| `work/` | Work notes index | `Index.md` (detailed MOC) |+| `work/active/` | **Current projects only** (1-3 files) | Move here when starting, move to archive when done |+| `work/archive/YYYY/` | Completed work organized by year | Grows over time |+| `work/incidents/` | Incident docs (main note + RCA + deep dive + drafts) | Per-incident grouping |+| `work/1-1/` | 1:1 meeting notes (accumulate weekly) | Named `<Person> YYYY-MM-DD.md` |+| `perf/` | Performance framework, brag doc | `Brag Doc.md` (index) |+| `perf/brag/` | Quarterly brag notes | One per quarter, e.g. `Q1 2025.md` |+| `perf/competencies/` | Atomic competency notes (link targets) | One note per competency |+| `perf/evidence/` | PR deep scans, data extracts for reviews | Named `<Person> PRs - <Period>.md` |+| `perf/<cycle>/` | Review cycle briefs + artifacts | Review briefs (private, manager, peer) |+| `brain/` | Claude's operational knowledge | `Memories.md`, `Key Decisions.md`, `Patterns.md`, `Gotchas.md`, `Skills.md`, `North Star.md` |+| `org/` | Organizational knowledge index | `People & Context.md` (MOC) |+| `org/people/` | Atomic person notes | One note per person |+| `org/teams/` | Team notes as graph nodes | One note per team |+| `reference/` | Codebase knowledge, architecture maps | Flow docs, architecture docs |+| `thinking/` | Scratchpad for drafts and reasoning | Named `YYYY-MM-DD-topic.md` |+| `templates/` | Obsidian templates | `Work Note.md`, `Decision Record.md`, etc. |## Obsidian CLI@@ -48,55 +76,90 @@ obsidian orphans # Unlinked notesA SessionStart hook automatically injects the vault file listing into context. No need to run `ls` or `find` -- you already know what files exist.-1. Read `claude/North Star.md` -- ground suggestions in current goals-2. Check `work/Index.md` -- see active projects and recent notes-3. Scan `claude/Memories.md` -- index of memory topics, then read relevant topic notes-4. `obsidian tasks daily todo` -- see pending items+1. Read `Home.md` -- vault entry point with embedded dashboards+2. Read `brain/North Star.md` -- ground suggestions in current goals+3. Check `work/Index.md` -- see active projects and recent notes+4. Scan `brain/Memories.md` -- index of memory topics, then read relevant topic notes+5. `obsidian tasks daily todo` -- see pending items### Ending a Substantial Session-Before wrapping up a session where meaningful work was done:+**When the user says "wrap up", "let's wrap", "wrapping up", or similar -- invoke `/wrap-up` automatically.** This runs a full review of the session.-1. Update `work/Index.md` if new notes or decisions were created-2. Update the relevant memory topic note (not `Memories.md` itself) with key learnings-3. Update `perf/Brag Doc.md` if wins or impact were achieved-4. Offer to update `claude/North Star.md` if goals shifted or new focus emerged-5. Verify all new notes link to at least one existing note-6. If work demonstrates competencies, add competency links to the work note's `## Related`+If `/wrap-up` is not invoked, at minimum do these before wrapping up:++1. **Archive completed projects**: `git mv` from `work/active/` to `work/archive/YYYY/`, update `status: completed` (or use `/project-archive`)+2. Update `work/Index.md` if new notes or decisions were created+3. Update the relevant brain topic note (`brain/Key Decisions.md`, `brain/Patterns.md`, `brain/Gotchas.md`) with key learnings+4. Update `org/People & Context.md` if org knowledge changed+5. Update `perf/Brag Doc.md` if wins or impact were achieved+6. Offer to update `brain/North Star.md` if goals shifted or new focus emerged+7. Verify all new notes link to at least one existing note (orphans are bugs)+8. If work demonstrates competencies, add competency links to the work note's `## Related`+9. Run `/vault-audit` if the session created many notesSkip steps that don't apply. The goal is transferring durable knowledge from conversation to vault state.### Thinking Workflow-Use `thinking/` for drafts, reasoning, and analysis before writing final notes. **Thinking notes are scratchpads, not storage.** They exist to help you reason — once the reasoning produces durable knowledge, promote it to proper notes and delete the scratchpad.+Use `thinking/` for drafts, reasoning, and analysis before writing final notes. **Thinking notes are scratchpads, not storage.** They exist to help you reason -- once the reasoning produces durable knowledge, promote it to proper notes and delete the scratchpad.1. Create a thinking note: `thinking/YYYY-MM-DD-descriptive-name.md`2. Use the Thinking Note template3. Reason through the problem, analyze options, draft content-4. Promote findings to atomic notes in the correct folder (not one monolith — one note per distinct concept)-5. Delete the thinking note — it served its purpose+4. Promote findings to atomic notes in the correct folder (not one monolith -- one note per distinct concept)+5. Delete the thinking note -- it served its purpose6. If the thinking process itself is worth preserving (unusual), keep it but link to the promoted notes### Creating Notes-1. **Always use YAML frontmatter** with at minimum `date`, `description` (~150 chars), `tags`, and type-specific fields.+1. **Always use YAML frontmatter** with at minimum `date`, `description` (~150 chars), `tags`, and type-specific fields. Work notes and incidents also need `quarter` (e.g., `Q1-2026`). Incidents need `ticket`, `severity`, `role`.2. **Use templates** from `templates/`. Fill `{{placeholders}}` with real values.-3. **Place files correctly**: work notes and decisions in `work/`, drafts in `thinking/`, perf content in `perf/`, competency notes in `perf/competencies/`, Claude context in `claude/`. Nothing in vault root except this file.+3. **Place files correctly**:+ - **Active** work notes, decisions, peer review prep -- `work/active/`+ - **Completed** work notes -- `work/archive/YYYY/` (by year)+ - Incident docs -- `work/incidents/`+ - 1:1 meeting notes -- `work/1-1/`+ - Performance content -- `perf/` (cycle subfolder for review briefs)+ - PR evidence -- `perf/evidence/`+ - Competency definitions -- `perf/competencies/`+ - People -- `org/people/`+ - Teams -- `org/teams/`+ - Claude operational context -- `brain/`+ - Codebase knowledge -- `reference/`+ - Drafts -- `thinking/`+ - Vault root: only `Home.md` and `CLAUDE.md`.4. **Name files descriptively.** Use the note title as filename.+### Note Types++| Type | Location | Naming | Key Sections |+|------|----------|--------|--------------|+| Work note | `work/active/` (then `archive/YYYY/` when done) | Descriptive title | Context, What/Why, Links, Related |+| Incident | `work/incidents/` | Ticket number or descriptive title | Context, Root Cause, Timeline, Impact, Analysis, Related |+| 1:1 note | `work/1-1/` | `<Person> YYYY-MM-DD.md` | Key Takeaways, Action Items, Quotes, What to Watch, Related |+| PR analysis | `perf/evidence/` | `<Person> PRs - <Period>.md` | PR Count, Projects, Quality, Growth, Full Table |+| Review brief | `perf/<cycle>/` | `H1 2026 Review Brief.md` | Arc, Impact, Competencies, Documentation Trail |+| Person note | `org/people/` | Full name | Role & Team, Relationship, Key Moments, Notes |+| Team note | `org/teams/` | Team name | Members, Scope, Interactions |+| Competency | `perf/competencies/` | Competency name | Definition, level criteria, Evidence (via backlinks) |+| Brain note | `brain/` | Topic name | Topic-specific content |+### Linking -- This Is Critical-**Graph principle**: This vault is a graph, not a wiki. A note's value comes from its connections, not its length. Link FROM evidence TO concepts -- never curate evidence lists inside concept notes. Let backlinks do the work.+**Graph-first, not folder-first.** Folders help browse in the sidebar. Links help discover through connections. Both matter, but links are the primary organizational tool.++**A note without links is a bug.** When creating a note, the FIRST thing to do after writing content is add wikilinks. Every new note must link to at least one existing note.-**Atomicity rule**: Before writing or appending to any note, ask: "Does this cover multiple distinct concepts that could be separate nodes?" If a note has or would have 3+ independent sections that don't need each other to make sense, split into atomic notes that link to each other. A graph of small, connected nodes is more useful than a document with internal headings. This applies equally to creating new notes AND growing existing ones — when an existing note outgrows its original scope, split it.+**Atomicity rule**: Before writing or appending to any note, ask: "Does this cover multiple distinct concepts that could be separate nodes?" If a note has or would have 3+ independent sections that don't need each other to make sense, split into atomic notes that link to each other.Note types have graph roles:-- **Evidence nodes** (work notes, decisions): add outbound links to concepts they demonstrate+- **Evidence nodes** (work notes, 1:1s, PR analyses): add outbound links to concepts they demonstrate- **Concept nodes** (competencies, patterns): stay definitional -- evidence arrives via backlinks-- **Index nodes** (Index, Brag Doc, Memories): actively curate links -- they're navigational--Every new note must link to at least one existing note. Proactively suggest connections.+- **Index nodes** (Index, Brag Doc, Memories, People & Context): actively curate links -- they're navigational+- **Person nodes** (org/people/): link to projects, teams, evidence. Receive backlinks from work notes.+Link syntax:- `[[Note Title]]` -- standard wikilink- `[[Note Title|display text]]` -- aliased link- `[[Note Title#Heading]]` -- deep link to section@@ -106,23 +169,23 @@ Every new note must link to at least one existing note. Proactively suggest conn#### When to Link- **Work note <-> Decision**: bidirectional links-- **Work note -> Competency**: in `## Related`, link to competencies demonstrated with a brief annotation (e.g., `[[Competency Name]] -- what was demonstrated`)+- **Work note -> Competency**: in `## Related`, link to competencies demonstrated+- **Work note -> Team**: in `## Related`, link to team(s) involved+- **Work note -> Person**: link people involved (especially in 1:1 notes)+- **Person -> PR analysis**: link to their evidence file if one exists- **Brag Doc -> Work note**: every entry links to evidence-- **Brag Doc -> Competency**: in `#### Competency Evidence` per quarter, link both competency and evidence work note-- **Review Template -> Competency + Work note**: per-competency assessment row with evidence links- **Memories -> Source**: every memory links to where it was learned-- **Skills -> Usage**: skills link to notes where used- **Index -> Everything**: `work/Index.md` links to all work notes-- **Thinking -> Final note**: thinking notes link to what they feed into, and vice versa- **North Star -> Projects**: active focus areas link to project work notes### Maintaining IndexesUpdate these when creating or archiving notes:-- **`work/Index.md`** -- add to Active Projects or Recent Notes, move completed to Archive, keep Decisions Log current-- **`claude/Memories.md`** -- index of memory topics. Add new memories to the relevant topic note, not here.-- **`claude/Skills.md`** -- register vault-specific workflows (not obsidian-skills -- those are in `.claude/skills/`)+- **`work/Index.md`** -- add to Active Projects or Recent Notes, move completed to Archive+- **`brain/Memories.md`** -- index of memory topics. Add new memories to the relevant topic note, not here.+- **`brain/Skills.md`** -- register vault-specific workflows and slash commands+- **`org/People & Context.md`** -- update when people, teams, or org structure changes- **`perf/Brag Doc.md`** -- log wins with links to evidence, add new quarters as needed### Decision Records@@ -130,7 +193,7 @@ Update these when creating or archiving notes:1. Create in `work/` using the Decision Record template2. Link from the work note(s) that led to the decision3. Add to the Decisions Log table in `work/Index.md`-4. If significant, note in the relevant memory topic note+4. If significant, note in `brain/Key Decisions.md`### Wins & Achievements@@ -138,7 +201,7 @@ When significant work is completed, add to `perf/Brag Doc.md` with links to the## North Star-`claude/North Star.md` is a living document of goals and focus areas.+`brain/North Star.md` is a living document of goals and focus areas.- **Read it** at the start of substantial sessions- **Reference it** when suggesting priorities or trade-offs@@ -149,27 +212,74 @@ When significant work is completed, add to `perf/Brag Doc.md` with links to theUse tags in frontmatter (not inline):-- **Type**: `work-note`, `decision`, `claude`, `perf`, `thinking`, `north-star`, `competency`+- **Type**: `work-note`, `decision`, `perf`, `thinking`, `north-star`, `competency`, `person`, `team`, `brain`- **Index**: `index`, `moc`- **Status** (frontmatter field): `active`, `completed`, `archived`, `proposed`, `accepted`, `deprecated`-- **Project**: as needed, e.g. `project/my-project`+- **Team** (frontmatter field on people + work notes): your team names, e.g. `Backend`, `Platform`, `Mobile`+- **Cycle** (frontmatter field on review-related notes): `h1-2026`, `h2-2025`, etc.+- **Person** (frontmatter field on evidence notes): full name of the person+- **Project**: as needed, e.g. `project/auth-refactor`++## Properties for Querying++Beyond tags, use these frontmatter properties to enable search and Bases views:++- `cycle: h1-2026` -- find all review material for a cycle+- `person: "Jane Smith"` -- find all evidence related to a person+- `team: Backend` -- find all notes related to a team+- `status: active` -- find active projects+- `quarter: Q1-2026` -- find all work for a quarter (used by Work Dashboard Base)+- `ticket: TICKET-123` -- find incident by ticket number+- `severity: high` -- incident severity+- `role: incident-lead` -- your role in an incident## Two Memory Systems| System | Location | Purpose ||--------|----------|---------|| **Claude Code memory** | `~/.claude/` | Auto-loaded. Workflow prefs, quick-recall. |-| **Vault memories** | `claude/Memories.md` + topic notes | Part of the graph. Rich, linked knowledge. |+| **Vault memories** | `brain/Memories.md` + topic notes | Part of the graph. Rich, linked knowledge. |Claude Code memory for session-level preferences. Vault memories for knowledge that benefits from linking and Obsidian browsing.+## Agent Guidelines++### Graph-First Thinking++- **Folders group by purpose, links group by meaning.** A note lives in ONE folder (its home) but links to MANY notes (its context).+- When creating a note, add wikilinks FIRST. A note without links is a bug.+- Prefer bidirectional links: if A links to B, B should link back to A (unless B is a concept node that receives backlinks passively).+- Before creating a new subfolder, ask: "Can I solve this with a tag, a property, or a link instead?" Folders are for browsing convenience, not for categorization.+- After every substantial session, verify new notes have at least one inbound link.++### Where to Put Things++- **Writing about a person?** -- `org/people/`+- **Writing about a team?** -- `org/teams/`+- **Writing about how the codebase works?** -- `brain/` (Patterns, Gotchas, Key Decisions)+- **Writing about what Claude should remember?** -- `brain/Memories.md` topic notes+- **Capturing a 1:1 meeting?** -- `work/1-1/`+- **Deep scanning PRs for review?** -- `perf/evidence/`+- **Creating review briefs?** -- `perf/<cycle>/`+- **Tracking active project work?** -- `work/active/`+- **Capturing an incident?** -- `work/incidents/` (use `/incident-capture`)++### Don't Mix Contexts++When capturing data from Slack, DMs, or meetings:+- **Project evidence** (PRs, technical decisions, delivery) -- goes to the relevant `work/` note+- **Review prep** (peer selection, manager strategy, brag framing) -- goes to review-related notes in `perf/` or `work/`+- **People dynamics** (feedback, relationships, career) -- goes to `org/people/` notes+- **Personal conversations** -- only capture if review-relevant; otherwise skip+## Rules- Never modify `.obsidian/` config files unless explicitly asked.- Preserve existing frontmatter when editing notes.-- Don't configure git hooks or auto-commit unless the user asks. Sync is handled outside Claude.+- Git sync is handled by `obsidian-git` -- don't configure git hooks or auto-commit.- When asked to "remember" something, write to the relevant memory topic note (not `Memories.md` itself) with a link to context.- Prefer Obsidian CLI over filesystem when Obsidian is running.-- **Always invoke Obsidian skills via the Skill tool** before doing vault work. Load `obsidian-markdown` when creating/editing `.md` files. Load `obsidian-cli` when running vault commands. Load `obsidian-bases` or `json-canvas` when working with those file types. Skills provide conventions and reference material — don't rely on memory alone.+- **Always invoke Obsidian skills via the Skill tool** before doing vault work. Load `obsidian-markdown` when creating/editing `.md` files. Load `obsidian-cli` when running vault commands. Load `obsidian-bases` or `json-canvas` when working with those file types.- Always check for and suggest connections between notes.-- Every note must have a `description` field (~150 chars). Claude fills this automatically. Enables scanning without reading full contents.+- Every note must have a `description` field (~150 chars). Claude fills this automatically.+- **Zero data loss**: when reorganizing, always use `git mv`. Never delete without explicit user confirmation.diff --git a/Home.md b/Home.mdnew file mode 100644index 0000000..cbce391--- /dev/null+++ b/Home.md@@ -0,0 +1,32 @@+---+description: "Vault entry point — embedded dashboards, quick links, current focus"+tags:+ - index+---++# Home++## Current Focus++![[North Star#Current Focus]]++## Active Work++![[Work Dashboard.base#Active Work]]++## Incidents++![[Incidents.base#All Incidents]]++## Quick Links++- [[Index|Work Notes]] | [[People & Context]] | [[Brag Doc]]+- [[Memories]] | [[North Star]] | [[Skills]]++## Recent 1:1s++![[1-1 History.base#All 1:1s]]++## People++![[People Directory.base#By Team]]diff --git a/README.md b/README.mdindex 650873c..a241b74 100644--- a/README.md+++ b/README.md@@ -1,116 +1,184 @@# Obsidian Mind-An Obsidian vault template designed for engineers who use Claude Code as a thinking partner. Your external brain for work notes, decisions, performance tracking, and persistent AI context.+An Obsidian vault template for engineers who use Claude Code as a thinking partner. Your external brain for work notes, decisions, performance tracking, and persistent AI context.## What This Is-A ready-to-use vault structure that makes Claude Code productive from session one. Every conversation builds on the last because Claude has persistent context in the vault — your goals, your decisions, your patterns, your wins.+A ready-to-use vault structure that makes Claude Code productive from session one. Every conversation builds on the last because Claude has persistent context in the vault -- your goals, your decisions, your patterns, your wins.++The vault is built around **wikilinks**. Every note links to related notes, building a knowledge graph that compounds over time. Claude maintains the indexes and links as notes are created, so the graph stays navigable without manual effort.++## What's New in V2++V2 is a significant expansion of the original template, adding structured workflows for performance tracking, incident management, and organizational knowledge.++- **`Home.md` dashboard** -- vault entry point with embedded Base views for active work, incidents, 1:1 history, and people+- **`bases/`** -- 7 centralized dynamic database views (Work Dashboard, Incidents, People Directory, 1-1 History, Review Evidence, Competency Map, Templates)+- **`work/active/` + `work/archive/`** -- explicit project lifecycle. Active projects live in `active/`, completed ones move to `archive/YYYY/`+- **`work/incidents/`** -- structured incident tracking with root cause analysis, timelines, and deep dives+- **`org/`** -- organizational knowledge with `org/people/` for person notes and `org/teams/` for team context+- **8 slash commands** -- `/peer-scan`, `/slack-scan`, `/capture-1on1`, `/vault-audit`, `/review-brief`, `/incident-capture`, `/project-archive`, `/wrap-up`+- **3 hooks** -- SessionStart (injects vault file listing), PreToolUse (validates frontmatter on writes), Stop (end-of-session checklist)+- **Enhanced `CLAUDE.md`** -- full session workflow, note type reference, graph-first linking rules, index maintenance, and agent guidelines## Quick Start-1. Clone this repo (or use it as a GitHub template)-2. Open the folder as an Obsidian vault-3. Enable the Obsidian CLI in Settings > Core plugins (requires Obsidian 1.12+)-4. Run `claude` in the vault directory-5. Fill in `claude/North Star.md` with your goals — this grounds every session+1. **Clone** this repo (or use it as a GitHub template)+2. **Open** the folder as an Obsidian vault+3. **Enable Bases** in Settings > Core plugins+4. **Enable the Obsidian CLI** in Settings > Core plugins (requires Obsidian 1.12+)+5. **Run** `claude` in the vault directory+6. **Fill in** `brain/North Star.md` with your current goals -- this grounds every session++Claude will read `CLAUDE.md` automatically and understand the full vault structure, conventions, and workflows.## Vault Structure```-work/ Work notes, project tracking, decision records-perf/ Brag doc, performance review templates-perf/competencies/ One note per competency (link targets for work notes)-claude/ Claude context: memories, skills, goals (North Star)-thinking/ Claude's scratchpad for drafts and reasoning-templates/ Obsidian templates with YAML frontmatter+Home.md Vault entry point -- embedded dashboards+CLAUDE.md Operating manual for Claude (loaded every session)++bases/ Centralized dynamic database views+ Work Dashboard.base Active and recent work+ Incidents.base All incidents with severity and status+ People Directory.base People organized by team+ 1-1 History.base Meeting notes timeline+ Review Evidence.base PR scans and evidence for reviews+ Competency Map.base Competency definitions and levels+ Templates.base Available note templates++work/ Work notes index+ Index.md Map of Content -- central hub for all work notes+ active/ Current projects (1-3 files)+ archive/ Completed work organized by year+ 2025/+ 2026/+ incidents/ Incident docs (main note + RCA + deep dive)+ 1-1/ 1:1 meeting notes (Person YYYY-MM-DD.md)++perf/ Performance framework+ Brag Doc.md Running log of wins, linked to evidence+ brag/ Quarterly brag notes (Q1 2025.md, etc.)+ competencies/ One note per competency (link targets)+ evidence/ PR deep scans and data extracts for reviews++brain/ Claude's operational knowledge+ Memories.md Index of memory topics+ Key Decisions.md Significant technical and architectural decisions+ Patterns.md Recurring patterns and approaches+ Gotchas.md Known pitfalls and workarounds+ Skills.md Custom workflows and slash command docs+ North Star.md Goals and focus areas (read every session)++org/ Organizational knowledge+ People & Context.md Index of people, teams, and org structure+ people/ One note per person+ teams/ One note per team++reference/ Codebase knowledge, architecture maps+thinking/ Scratchpad for drafts and reasoning+templates/ Obsidian templates with YAML frontmatter```-### Key Files--| File | Purpose |-|------|---------|-| `CLAUDE.md` | Operating manual for Claude — loaded automatically every session |-| `claude/North Star.md` | Your goals and focus areas — grounds Claude's suggestions |-| `claude/Memories.md` | Persistent knowledge Claude retains across sessions |-| `claude/Skills.md` | Custom workflows and reusable prompts |-| `work/Index.md` | Map of Content — central hub for all work notes |-| `perf/Brag Doc.md` | Running log of wins, linked to evidence |+## Slash Commands-## What's Included+| Command | Purpose |+|---------|---------|+| `/peer-scan` | Deep scan a peer's GitHub PRs for review prep |+| `/slack-scan` | Deep scan Slack channels/DMs for evidence |+| `/capture-1on1` | Capture 1:1 meeting transcript into structured vault note |+| `/vault-audit` | Audit indexes, links, orphans, stale context |+| `/review-brief` | Generate review brief (manager or peer version) |+| `/incident-capture` | Capture incident from Slack channels/DMs into structured vault notes |+| `/project-archive` | Move completed project from active/ to archive/, update indexes |+| `/wrap-up` | Full session review -- verify notes, indexes, links, suggest improvements |-### Obsidian Skills+Commands are defined in `.claude/commands/` and can be invoked directly in Claude Code with the slash prefix.-[kepano/obsidian-skills](https://github.com/kepano/obsidian-skills) pre-installed in `.claude/skills/`:+## Bases-- **obsidian-markdown** — Teaches Claude Obsidian-flavored markdown (wikilinks, embeds, callouts, properties)-- **obsidian-cli** — CLI commands for vault operations-- **json-canvas** — Visual canvas file creation-- **obsidian-bases** — Database-style base files-- **defuddle** — Web page to markdown extraction+The vault includes 7 Obsidian Base files in `bases/` that provide dynamic database views over your vault content.-### SessionStart Hook+| Base | What It Shows |+|------|---------------|+| Work Dashboard | Active projects filtered by status and quarter |+| Incidents | All incidents with severity, role, and ticket number |+| People Directory | People organized by team with roles |+| 1-1 History | Timeline of all 1:1 meeting notes |+| Review Evidence | PR scans and evidence linked to people and cycles |+| Competency Map | Competency definitions with level criteria |+| Templates | Available note templates for quick reference |-A hook in `.claude/settings.json` injects the vault file listing into every Claude session. Claude immediately knows what exists — no wasted turns on discovery.+Bases require the Bases core plugin (Obsidian 1.8+). They query frontmatter properties like `status`, `quarter`, `team`, `person`, and `cycle`.-### Templates+## Hooks-Four templates with YAML frontmatter including a `description` field for progressive disclosure:+Three hooks in `.claude/settings.json` automate vault hygiene:-- **Work Note** — date, description, project, status, tags-- **Decision Record** — date, description, status (proposed/accepted/deprecated), tags-- **Thinking Note** — date, description, context, tags-- **Competency Note** — date, description, current-level, target-level, proficiency table+### SessionStart+Runs on startup, resume, and clear. Injects the full vault file listing into Claude's context so it immediately knows what exists -- no wasted turns on file discovery. Also prints a quick-reference reminder of vault conventions.-### Session Workflow+### PreToolUse+Triggers before any file write. Reminds Claude to include proper frontmatter (`date`, `quarter`, `description`, `tags`), add at least one wikilink, and place the file in the correct folder per `CLAUDE.md`.-CLAUDE.md defines start and end workflows:+### Stop+Triggers at the end of every session. Prints a checklist: archive completed projects, update indexes, verify new notes are linked, and optionally run `/vault-audit`.-- **Start**: Read North Star, check Index, scan Memories-- **End**: Update indexes, capture learnings, log wins+## Key Files-## How It Works+Start by customizing these files for your situation:-The vault is built around **wikilinks**. Every note links to related notes, building a graph that compounds over time:+| File | What to Do |+|------|------------|+| `brain/North Star.md` | Write your current goals and focus areas |+| `perf/competencies/` | Create notes for your organization's competency framework |+| `org/People & Context.md` | Add your team, manager, and key collaborators |+| `org/teams/` | Create notes for teams you work with |+| `work/Index.md` | Will grow as you add work notes |+| `perf/Brag Doc.md` | Will grow as you log wins |-- Work notes link to decisions that came from them-- The brag doc links to work notes as evidence-- Memories link to where they were learned-- The thinking folder links drafts to their final notes--Claude maintains the indexes (`work/Index.md`, `claude/Memories.md`, `perf/Brag Doc.md`) as notes are created, so the graph stays navigable.--## Performance Graph+## Customization-The vault includes a performance tracking system built on the same graph principles:+This is a starting point, not a straitjacket. Adapt it to how you work:-- **Competency notes** in `perf/competencies/` represent skills from your organization's competency framework-- **Work notes** link to competencies in `## Related` — annotated with what was demonstrated-- **Brag Doc** aggregates evidence per quarter in `#### Competency Evidence` sections-- **Review Template** structures self-assessment around Impact, Competencies, and Principles-- **Backlinks accumulate automatically** — review prep becomes reading the backlinks panel on each competency note+- **Teams**: Replace the generic team names in `CLAUDE.md` tags convention with your actual team names+- **Competencies**: Create notes in `perf/competencies/` matching your organization's framework (IC3/IC4/IC5 levels, or whatever your org uses)+- **Properties**: Add custom frontmatter properties and update `CLAUDE.md` Properties for Querying section+- **Templates**: Modify templates in `templates/` or add new ones+- **Slash commands**: Edit commands in `.claude/commands/` to match your tools (GitHub org, Slack workspace, etc.)+- **Bases**: Customize the `.base` files in `bases/` to add filters, columns, or new views+- **Reference**: Add codebase architecture docs, flow diagrams, or domain knowledge to `reference/`+- **Review cycles**: Create `perf/<cycle>/` folders as review periods come up (e.g., `perf/h1-2026/`)-To get started: create competency notes using the `Competency Note` template, then link your work notes to them as you complete projects.+## Migrating from V1-## Customization+If you used the original Obsidian Mind template (V1), here is what changed:-This is a starting point, not a straitjacket. Adapt it to how you work:+1. **Rename `claude/` to `brain/`** -- `git mv claude/ brain/`+2. **Create new folders**: `bases/`, `org/`, `org/people/`, `org/teams/`, `work/active/`, `work/archive/`, `work/incidents/`, `work/1-1/`, `perf/brag/`, `perf/evidence/`, `reference/`+3. **Move active work notes** from `work/` to `work/active/`+4. **Move completed work notes** from `work/` to `work/archive/YYYY/`+5. **Create `Home.md`** in vault root with embedded Base views+6. **Copy Base files** from the V2 template into `bases/`+7. **Add new commands**: copy missing `.claude/commands/` files from V2+8. **Update `CLAUDE.md`**: replace with the V2 version, then re-add any custom conventions+9. **Update `.claude/settings.json`**: add PreToolUse and Stop hooks from V2+10. **Update tags**: `claude` tag becomes `brain`, add `person` and `team` tags-- Add folders for your domain (e.g., `meetings/`, `1on1s/`, `research/`)-- Create new templates in `templates/`-- Add custom skills in `claude/Skills.md`-- Extend the CLAUDE.md with your own conventions+All V1 content is preserved -- V2 only adds structure around it.## Requirements-- [Obsidian](https://obsidian.md) 1.12+ (for CLI support)+- [Obsidian](https://obsidian.md) 1.12+ (for CLI support and Bases)- [Claude Code](https://claude.ai/claude-code)+- Bases core plugin enabled (Settings > Core plugins)- Git (for version history; sync via git, [Obsidian Sync](https://obsidian.md/sync), or your preferred method)## Design Influences-- [kepano/obsidian-skills](https://github.com/kepano/obsidian-skills) — Official Obsidian agent skills-- [James Bedford](https://x.com/jameesy) — Vault structure philosophy, separation of AI-generated content-- [arscontexta](https://github.com/agenticnotetaking/arscontexta) — Progressive disclosure via description fields, session hooks, kernel primitives+- [kepano/obsidian-skills](https://github.com/kepano/obsidian-skills) -- Official Obsidian agent skills+- [James Bedford](https://x.com/jameesy) -- Vault structure philosophy, separation of AI-generated content+- [arscontexta](https://github.com/agenticnotetaking/arscontexta) -- Progressive disclosure via description fields, session hooks, kernel primitives## Licensediff --git a/bases/1-1 History.base b/bases/1-1 History.basenew file mode 100644index 0000000..cced48c--- /dev/null+++ b/bases/1-1 History.base@@ -0,0 +1,18 @@+filters:+ and:+ - file.inFolder("work/1-1")+ - file.ext == "md"+properties:+ file.name:+ displayName: Meeting+ date:+ displayName: Date+ description:+ displayName: Key Takeaways+views:+ - type: table+ name: All 1:1s+ order:+ - file.name+ - date+ - descriptiondiff --git a/bases/Competency Map.base b/bases/Competency Map.basenew file mode 100644index 0000000..b314354--- /dev/null+++ b/bases/Competency Map.base@@ -0,0 +1,21 @@+filters:+ and:+ - file.hasTag("competency")+ - file.ext == "md"+ - '!file.inFolder("templates")'+formulas:+ evidence_count: file.backlinks.length+properties:+ file.name:+ displayName: Competency+ formula.evidence_count:+ displayName: Evidence Count+ description:+ displayName: Definition+views:+ - type: table+ name: Competencies+ order:+ - file.name+ - formula.evidence_count+ - descriptiondiff --git a/bases/Incidents.base b/bases/Incidents.basenew file mode 100644index 0000000..bbcfa9e--- /dev/null+++ b/bases/Incidents.base@@ -0,0 +1,32 @@+filters:+ and:+ - file.hasTag("incident")+ - file.ext == "md"+formulas:+ duration_info: if(date, (today() - date(date)).days.toString() + "d ago", "")+properties:+ file.name:+ displayName: Incident+ irp:+ displayName: IRP Ticket+ severity:+ displayName: Severity+ role:+ displayName: Your Role+ status:+ displayName: Status+ date:+ displayName: Date+ description:+ displayName: Summary+views:+ - type: table+ name: All Incidents+ order:+ - file.name+ - date+ - irp+ - severity+ - role+ - status+ - descriptiondiff --git a/bases/People Directory.base b/bases/People Directory.basenew file mode 100644index 0000000..522bcad--- /dev/null+++ b/bases/People Directory.base@@ -0,0 +1,31 @@+filters:+ and:+ - file.hasTag("person")+ - file.ext == "md"+properties:+ file.name:+ displayName: Name+ team:+ displayName: Team+ title:+ displayName: Role+ description:+ displayName: Context+views:+ - type: table+ name: By Team+ groupBy:+ property: team+ direction: ASC+ order:+ - file.name+ - title+ - team+ - description+ - type: table+ name: All People+ order:+ - file.name+ - title+ - team+ - descriptiondiff --git a/bases/Review Evidence.base b/bases/Review Evidence.basenew file mode 100644index 0000000..11e6b18--- /dev/null+++ b/bases/Review Evidence.base@@ -0,0 +1,38 @@+filters:+ and:+ - file.inFolder("perf/evidence")+ - file.ext == "md"+properties:+ file.name:+ displayName: Evidence+ person:+ displayName: Person+ cycle:+ displayName: Cycle+ date:+ displayName: Date+ description:+ displayName: Summary+views:+ - type: table+ name: By Cycle+ groupBy:+ property: cycle+ direction: DESC+ order:+ - file.name+ - person+ - cycle+ - date+ - description+ - type: table+ name: By Person+ groupBy:+ property: person+ direction: ASC+ order:+ - file.name+ - person+ - cycle+ - date+ - descriptiondiff --git a/bases/Templates.base b/bases/Templates.basenew file mode 100644index 0000000..81875bd--- /dev/null+++ b/bases/Templates.base@@ -0,0 +1,18 @@+filters:+ and:+ - file.inFolder("templates")+ - file.ext == "md"+properties:+ file.name:+ displayName: Template+ description:+ displayName: Purpose+ file.mtime:+ displayName: Last Modified+views:+ - type: table+ name: All Templates+ order:+ - file.name+ - description+ - file.mtimediff --git a/bases/Work Dashboard.base b/bases/Work Dashboard.basenew file mode 100644index 0000000..020c5f4--- /dev/null+++ b/bases/Work Dashboard.base@@ -0,0 +1,61 @@+filters:+ and:+ - file.hasTag("work-note")+ - file.ext == "md"+ - '!file.inFolder("templates")'+formulas:+ quarter: if(date, "Q" + ((date(date).month - 1) / 3 + 1).floor().toString() + " " + date(date).year.toString(), "")+ age_days: if(date, (today() - date(date)).days, "")+properties:+ file.name:+ displayName: Note+ status:+ displayName: Status+ project:+ displayName: Project+ formula.quarter:+ displayName: Quarter+ date:+ displayName: Date+ description:+ displayName: Summary+views:+ - type: table+ name: Active Work+ filters:+ and:+ - status == "active"+ groupBy:+ property: project+ direction: ASC+ order:+ - file.name+ - date+ - project+ - formula.quarter+ - description+ - type: table+ name: All by Quarter+ groupBy:+ property: formula.quarter+ direction: DESC+ order:+ - file.name+ - date+ - status+ - project+ - description+ - type: table+ name: Completed+ filters:+ and:+ - status == "completed"+ groupBy:+ property: formula.quarter+ direction: DESC+ order:+ - file.name+ - date+ - project+ - formula.quarter+ - descriptiondiff --git a/brain/Gotchas.md b/brain/Gotchas.mdnew file mode 100644index 0000000..13c92c7--- /dev/null+++ b/brain/Gotchas.md@@ -0,0 +1,11 @@+---+description: "Things that have bitten before and will bite again — pitfalls, edge cases, and testing traps"+tags:+ - brain+---++# Gotchas++Things that have bitten before and will bite again.++-diff --git a/brain/Key Decisions.md b/brain/Key Decisions.mdnew file mode 100644index 0000000..77b60c1--- /dev/null+++ b/brain/Key Decisions.md@@ -0,0 +1,11 @@+---+description: "Architectural and workflow decisions worth recalling across sessions — each links to its source work note"+tags:+ - brain+---++# Key Decisions++Architectural or workflow decisions worth recalling. Link to the full [[Decision Record]] when one exists.++-diff --git a/brain/Memories.md b/brain/Memories.mdnew file mode 100644index 0000000..2917eca--- /dev/null+++ b/brain/Memories.md@@ -0,0 +1,21 @@+---+description: "Index of memory topics — key decisions, patterns, gotchas, people context"+tags:+ - brain+ - index+---++# Memories++Persistent context and knowledge retained across sessions. Each topic lives in its own note — follow the links.++- [[Key Decisions]] — architectural and workflow decisions worth recalling+- [[Patterns]] — recurring patterns and conventions discovered across work+- [[Gotchas]] — things that have bitten before and will bite again+- [[People & Context]] — org structure, teams, review history, dynamics+- [[North Star]] — living goals document, read at session start+- [[Skills]] — custom slash commands and workflows++## Recent Context++-diff --git a/claude/North Star.md b/brain/North Star.mdsimilarity index 84%rename from claude/North Star.mdrename to brain/North Star.mdindex 4691920..6ca0521 100644--- a/claude/North Star.md+++ b/brain/North Star.md@@ -1,7 +1,8 @@---date:+description: "Living document of goals, focus areas, and aspirations — read at session start, updated when direction shifts"tags:- - claude+ - brain- north-staraliases:- Goalsdiff --git a/brain/Patterns.md b/brain/Patterns.mdnew file mode 100644index 0000000..633faf0--- /dev/null+++ b/brain/Patterns.md@@ -0,0 +1,11 @@+---+description: "Recurring patterns and conventions discovered across work — architecture, naming, tooling, and implementation patterns"+tags:+ - brain+---++# Patterns++Recurring patterns discovered across work.++-diff --git a/brain/Skills.md b/brain/Skills.mdnew file mode 100644index 0000000..7f9fee3--- /dev/null+++ b/brain/Skills.md@@ -0,0 +1,50 @@+---+description: "Vault-specific workflows and slash commands — reusable patterns for review prep, project tracking, and vault maintenance"+tags:+ - brain+ - index+---++# Skills++Custom slash commands and reusable workflows. Defined in `.claude/commands/`.++## Slash Commands++| Command | Purpose |+|---------|---------|+| `/peer-scan` | Deep scan a peer's GitHub PRs for review prep |+| `/slack-scan` | Deep scan Slack channels/DMs for evidence |+| `/capture-1on1` | Capture 1:1 meeting transcript into structured vault note |+| `/vault-audit` | Audit indexes, links, orphans, stale context |+| `/review-brief` | Generate review brief (manager or peer version) |+| `/incident-capture` | Capture incident from Slack channels/DMs into structured vault notes |+| `/project-archive` | Move completed project from active/ to archive/, update indexes |+| `/wrap-up` | Full session review — verify notes, indexes, links, suggest improvements |++## Usage Notes++- `/peer-scan` works best when launched as parallel agents (one per person)+- `/slack-scan` should be run AFTER `/peer-scan` to add context beyond code+- `/capture-1on1` handles transcripts, raw notes, or summaries+- `/vault-audit` should be run at the end of substantial sessions+- `/review-brief` needs the private brief to exist first — it generates filtered versions+- `/incident-capture` takes Slack URLs and produces structured incident documentation+- `/project-archive` handles the active/ → archive/ move with index updates+- `/wrap-up` is auto-triggered when you say "wrap up" — runs full session review++## Workflow: Full Review Cycle Prep++1. **`/review-brief manager`** — generate the manager context transfer doc+2. **`/review-brief peers`** — generate the peer context transfer doc+3. **`/peer-scan`** (parallel, one per peer) — deep scan each peer's PRs+4. **`/slack-scan`** — scan relevant channels for your own evidence + peer context+5. **`/capture-1on1`** — capture the review 1:1 with your manager+6. **`/vault-audit`** — tidy up after all the new data++## Workflow: Project Ramp-Up++1. **`/slack-scan`** — scan project channels for history and decisions+2. **`/peer-scan`** (if needed) — understand what teammates have already built+3. Create work note from gathered context+4. **`/vault-audit`** — ensure everything links properlydiff --git a/claude/Memories.md b/claude/Memories.mddeleted file mode 100644index 72aa45a..0000000--- a/claude/Memories.md+++ /dev/null@@ -1,29 +0,0 @@-----tags:- - claude- - index------# Claude Memories--Persistent context and knowledge that Claude retains across sessions. Each entry links to the source note when applicable -- follow backlinks in Obsidian to see where knowledge came from.--## Key Decisions-Architectural or workflow decisions worth recalling. Link to the full [[Decision Record]] when one exists.-----## Patterns & Conventions-Recurring patterns discovered across work.-----## Gotchas-Things that have bitten before and will bite again.-----## People & Context-Key context about teams, stakeholders, or collaborators.---diff --git a/claude/Skills.md b/claude/Skills.mddeleted file mode 100644index f1cc461..0000000--- a/claude/Skills.md+++ /dev/null@@ -1,24 +0,0 @@-----tags:- - claude- - index------# Claude Skills--Custom skills, prompts, and reusable workflows. Each skill can link to work notes where it was used or refined.--## Active Skills-Named workflows Claude can execute on request.-----## Prompt Snippets-Reusable prompt fragments for common tasks.-----## Workflows-Multi-step processes. Link to the notes where these were developed.---diff --git a/org/People & Context.md b/org/People & Context.mdnew file mode 100644index 0000000..3d6b734--- /dev/null+++ b/org/People & Context.md@@ -0,0 +1,42 @@+---+description: "Org context, teams, stakeholders, review history, and dynamics"+tags:+ - org+ - index+---++# People & Context++Key context about teams, stakeholders, or collaborators. See also [[Memories]] for session context and [[Skills]] for workflows.++## Role & Org++-++## Teams++Team notes in `org/teams/`:++-++## Performance Reviews++-++## Manager Timeline++-++## People++Individual notes in `org/people/`:++-++## Review Goals++-++## Recurring Growth Themes++-diff --git a/org/people/.gitkeep b/org/people/.gitkeepnew file mode 100644index 0000000..e69de29diff --git a/org/teams/.gitkeep b/org/teams/.gitkeepnew file mode 100644index 0000000..e69de29diff --git a/perf/Brag Doc.md b/perf/Brag Doc.mdindex ff3ab6e..980f7a7 100644--- a/perf/Brag Doc.md+++ b/perf/Brag Doc.md@@ -1,4 +1,5 @@---+description: "Index of quarterly brag notes — impact, competency evidence, technical growth, feedback per quarter"tags:- perf- index@@ -6,23 +7,17 @@ tags:# Brag Doc-A running log of impact, wins, and growth. Link to work notes as evidence -- backlinks make perf reviews easy.+A running log of impact, wins, and growth. Each quarter is its own note — open the one you need.## Current Year-### Q1+| Quarter | Highlights | Review |+|---------|-----------|--------|+| | | |-#### Competency Evidence--+## How This Works-#### Impact & Deliverables----#### Technical Growth----#### Collaboration & Leadership----#### Feedback Received--+- Each quarter note has: Competency Evidence, Impact & Deliverables, Technical Growth, Collaboration, Feedback+- Competency evidence links both the competency and the work note+- Review prep: open the quarter(s) covered by the review period, follow the links+- Backlinks from quarterly notes accumulate on competency notes and work notes automaticallydiff --git a/perf/brag/.gitkeep b/perf/brag/.gitkeepnew file mode 100644index 0000000..e69de29diff --git a/perf/evidence/.gitkeep b/perf/evidence/.gitkeepnew file mode 100644index 0000000..e69de29diff --git a/reference/.gitkeep b/reference/.gitkeepnew file mode 100644index 0000000..e69de29diff --git a/perf/Review Template.md b/templates/Review Template.mdsimilarity index 100%rename from perf/Review Template.mdrename to templates/Review Template.mddiff --git a/work/1-1/.gitkeep b/work/1-1/.gitkeepnew file mode 100644index 0000000..e69de29diff --git a/work/Index.md b/work/Index.mdindex e674c64..39f30ec 100644--- a/work/Index.md+++ b/work/Index.md@@ -1,4 +1,5 @@---+description: "Central map of all work notes — active projects, completed work by quarter, decisions log"tags:- index- moc@@ -6,15 +7,39 @@ tags:# Work Notes-Central map of content. All work notes and decisions link back here.+Central map of content. All work notes and decisions link back here. For quick navigation, use [[Home]] or open `bases/Work Dashboard.base`.++**Folder structure**: `active/` = current projects, `archive/` = completed (by year), `incidents/` = incident docs, `1-1/` = meetings.++## Incidents++Incident docs live in `work/incidents/`. See `Incidents.base` for overview.++-## Active Projects-Projects currently in progress. Each links to its related work notes.--## Recent Notes+## Review Prep+-++## Recently Completed++-++## Completed++### Current Quarter+-++### Previous Quarters+-++## Reference++-## Decisions Log@@ -27,6 +52,5 @@ Projects currently in progress. Each links to its related work notes.-## Archive-Completed or abandoned projects.-diff --git a/work/active/.gitkeep b/work/active/.gitkeepnew file mode 100644index 0000000..e69de29diff --git a/work/archive/.gitkeep b/work/archive/.gitkeepnew file mode 100644index 0000000..e69de29diff --git a/work/incidents/.gitkeep b/work/incidents/.gitkeepnew file mode 100644index 0000000..e69de29