[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M76W582J6PD3TTXZXFA2HVQJ":3,"skill-same:01M76W582J6PD3TTXZXFA2HVQJ":47},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":25,"commitSha":26,"skillDirectory":27,"forksCount":14,"repositoryHtmlUrl":28,"repositoryHomepageUrl":6,"ownerHtmlUrl":29,"ownerBlogUrl":30,"ownerDescription":31,"updatedAt":32,"install":33,"resources":46},"01M76W582J6PD3TTXZXFA2HVQJ","week-review",null,"Review week's progress with concrete accomplishments (not fake percentages), pattern detection, and goal tracking.","hemory\u002Famp","hemory","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F21179881?v=4","01M6D31HFJ391DN8N0JP8AQGFE","Hemory Phifer","01M758Z6KK3NGD8654Y5CT668C",0,"MIT","https:\u002F\u002Fgithub.com\u002Fhemory\u002Famp\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fweek-review\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: week-review\ndescription: Review week's progress with concrete accomplishments (not fake percentages), pattern detection, and goal tracking.\ncontext: fork\n---\n\n## Purpose\n\nCreate a synthesis of the week reviewing activity, progress, and what was accomplished. **Uses concrete metrics, not vague percentages.**\n\n---\n\n## Data Sources\n\n### 1. Task Progress\n- `03-Tasks\u002FTasks.md` — Task completion status\n- `02-Week_Priorities\u002FWeek_Priorities.md` — Weekly priorities\n\n### 2. Project Activity\n- `04-Projects\u002F**\u002F*.md` — Modified this week\n\n### 3. Meetings & People\n- `00-Inbox\u002FMeetings\u002F*.md` — Meeting notes from this week\n- `People\u002F**\u002F*.md` — Person pages updated\n\n### 4. Learnings\n- `06-Resources\u002FLearnings\u002F**\u002F*.md` — Explicit learnings\n- `System\u002FSession_Learnings\u002F*.md` — Auto-captured session learnings\n\n### 5. Daily Reviews\n- `07-Archives\u002FReviews\u002FYYYY-MM-DD - Daily Review.md` — This week's reviews (fall back to `Daily_Review_YYYY-MM-DD.md` for legacy naming)\n\n### 6. Journals (If Enabled)\n- `00-Inbox\u002FJournals\u002FYYYY\u002FMM-Month\u002F` — Morning\u002Fevening journals\n\n### 7. Slack Channels (If Enabled)\n- Configured channels from `System\u002Fuser-profile.yaml` > `slack_intel`\n- Pulled via `gh slack-aggregator review` with weekly date range\n\n---\n\n## Analysis Process\n\n### 1. Weekly Priority Completion (Concrete, Not Percentages)\n\n**Don't say:** \"Goal X went from 40% to 55%\"\n**Do say:** \"You completed 2 of 3 weekly priorities\"\n\n```\nUse: get_week_progress()\n```\n\nFor each weekly priority:\n- **Complete:** ✅ What was the deliverable? When did you finish?\n- **In Progress:** 🔄 What specifically got done? What's left?\n- **Not Started:** ❌ Why? Should it carry forward?\n\n**Surface concrete accomplishments:**\n\n> \"**This week's priorities:**\n> \n> 1. ✅ **Ship pricing page** — Complete (pushed to prod Wednesday)\n>    - Deliverable: New pricing page live at \u002Fpricing\n>    - Tasks completed: 5 of 5\n> \n> 2. 🔄 **Write Q1 strategy doc** — 60% complete\n>    - Done: Outline, competitive analysis, recommendations\n>    - Remaining: Executive summary, financial projections\n>    - 2 tasks left\n> \n> 3. ❌ **Customer interviews** — Not started\n>    - Reason: Calendar was too stacked\n>    - Recommendation: Carry to next week with protected time\"\n\n### 1.5 Semantic Goal-to-Work Mapping (if QMD available)\n\n**Check if semantic search is available** by looking for `qmd` in PATH.\n\nIf available, enhance the weekly priority review with meaning-based analysis:\n\n1. **Auto-detect goal contributions:** For each completed task this week, search:\n   ```\n   qmd query \"task title\u002Fdescription\" --limit 3\n   ```\n   against quarterly goals. Catch tasks that advanced goals without explicit links.\n   - Example: \"Built customer health dashboard\" semantically matches goal \"Improve NPS tracking\" — different words, same work.\n\n2. **Cross-priority connections:** Search for work that bridges multiple priorities:\n   ```\n   qmd query \"priority 1 description\" --limit 5\n   ```\n   Surface tasks that contributed to more than one priority.\n\n3. **Thematic patterns:** Search for recurring themes across the week's work:\n   ```\n   qmd query \"common theme from meetings\u002Ftasks\" --limit 5\n   ```\n   Detect patterns like \"most of your work this week clustered around customer retention\" even when tasks used different terminology.\n\n**Integration:** Merge findings into the Quarterly Goals table — add a \"Hidden contributions\" row for semantically-detected but not explicitly-linked work. Only show genuinely new connections, not things already captured by keyword matching.\n\n**If QMD unavailable:** Skip silently. Task completion stats still work fine.\n\n### 2. Task Completion Stats (Concrete Numbers)\n\nScan `03-Tasks\u002FTasks.md` for completion timestamps from this week:\n- Count tasks completed (look for `✅ YYYY-MM-DD` in date range)\n- Count tasks added mid-week\n- Count tasks carried over\n\n**Surface:**\n\n> \"**Tasks this week:**\n> - Completed: 14 tasks\n> - Added mid-week: 6 tasks (scope creep?)\n> - Carried over: 3 tasks\n> \n> **Completion rate:** 82% (14 of 17 planned)\"\n\n### 3. Quarterly Goals Progress (Concrete Milestones)\n\n**Don't use fake percentages.** Use milestone counts and specific accomplishments.\n\n```\nUse: get_quarterly_goals()\nUse: get_goal_status(goal_id) for each goal\n```\n\nFor each goal:\n- Milestones completed this week\n- Total milestones done vs. total\n- Weeks since last milestone\n- Specific accomplishments that moved the goal\n\n> \"**Quarterly Goals Progress:**\n> \n> | Goal | Milestones | This Week | Status |\n> |------|------------|-----------|--------|\n> | Launch v2.0 | 3 of 5 | +1 (Pricing page shipped) | On track |\n> | Improve NPS | 1 of 4 | No change | ⚠️ Stalled (3 weeks) |\n> | Team Capacity | 2 of 3 | No change | On track |\n> \n> **Goal 1** advanced because you completed Priority 1.\n> **Goal 2** needs attention — no linked work completed this week.\"\n\n### 4. Daily Completion Rate Trend (NEW)\n\nIf daily reviews exist, calculate completion trends:\n\n> \"**Daily plan completion this week:**\n> \n> | Day | Planned | Done | Rate |\n> |-----|---------|------|------|\n> | Mon | 3 | 2 | 67% |\n> | Tue | 3 | 3 | 100% |\n> | Wed | 3 | 2 | 67% |\n> | Thu | 3 | 1 | 33% |\n> | Fri | 3 | 2 | 67% |\n> \n> **Week average:** 67%\n> **Pattern:** Thursday was rough (too many meetings?)\"\n\n### 5. Meeting Analysis\n\nReview meeting notes from the week:\n- Meetings held\n- Key decisions\n- Action items created\n- Follow-ups that might have slipped\n\n### 5.5 Commitment Health Analysis (NEW)\n\nIf ScreenPipe and Commitment Detection are available, show aggregate stats:\n\n```\nUse: get_commitment_stats(\n    start_date=\"YYYY-MM-DD\",  # Monday of this week\n    end_date=\"YYYY-MM-DD\"     # Today\n)\n```\n\n**Surface to user:**\n\n> \"📊 **Commitment Health This Week**\n>\n> **Detected across apps:** 12 potential commitments\n> **Already had tasks:** 7 (58%)\n> **Created from prompts:** 3\n> **Dismissed as handled:** 2\n>\n> **Apps with most uncaptured asks:**\n> 1. Slack - 5 items\n> 2. Email - 4 items\n> 3. Notion - 3 items\n>\n> **People who asked most of you:**\n> 1. Sarah Chen - 4 asks\n> 2. Product team - 3 asks\n>\n> 💡 *Consider: Check Slack more frequently for asks, or run `\u002Fcommitment-scan` mid-week*\"\n\n**If no commitment data:**\nSkip this section silently (user may not have ScreenPipe or commitment detection enabled).\n\n### 5.8 Email Communication Stats (if Gmail connected)\n\nCheck `System\u002Fintegrations\u002Fconfig.yaml` for `google-workspace.enabled: true`.\n\nIf enabled and Google Workspace MCP is healthy:\n- **Emails sent this week** — count of sent messages in the review period\n- **Average response time** — how quickly you replied to incoming emails\n- **Threads still open** — conversations with no resolution (back-and-forth still active)\n- **Follow-up detection** — emails waiting > 48h for a reply from you or from others\n\nSurface in the review:\n\n> \"**Email this week:**\n>\n> | Metric | Value |\n> |--------|-------|\n> | Emails sent | 47 |\n> | Avg response time | 3.2 hours |\n> | Open threads | 12 |\n> | Awaiting your reply (> 48h) | 3 |\n>\n> **Observation:** You have 3 emails waiting for replies longer than 48 hours. Consider clearing those early next week.\"\n\nIf unhealthy or not enabled: skip this section silently.\n\n### 5.9 Slack Intelligence (if configured)\n\nCheck `System\u002Fuser-profile.yaml` > `slack_intel.enabled`.\n\nIf enabled and `gh slack-aggregator` is installed, run the `\u002Fslack-intel` skill workflow:\n\n1. Pull each configured channel using `gh slack-aggregator review --channel \u003Curl> --from \u003Cmonday> --to \u003Ctoday>`\n2. Analyze exported markdown against active projects, people, and pillars\n3. Categorize findings: 🎯 Action Items, 📋 Project Context, 👤 People Signals, 📊 Org Intel\n4. Present findings to the user for confirmation and routing\n5. Route confirmed items to tasks, project notes, person pages\n6. Save summary to `00-Inbox\u002FSlack_Intel\u002FWeek_of_YYYY-MM-DD.md`\n7. Clean up temporary export files\n\n**Full workflow details:** See `.claude\u002Fskills\u002Fslack-intel-custom\u002FSKILL.md`\n\nSurface in the review:\n\n> \"📡 **Slack Intelligence**\n>\n> **Channels scanned:** 3\n> **Items found:** X action items, X project updates, X people signals\n>\n> [Categorized findings presented for user confirmation]\"\n\nAction items from Slack Intel feed into the \"Next Week\" suggestions section.\n\nIf not enabled or aggregator not installed: skip this section silently.\n\n### 6. Learning Compilation & Pattern Detection\n\nReview `System\u002FSession_Learnings\u002F` files from this week:\n\n**Pattern Detection:**\n- **Recurring issues:** Same mistake 2+ times? Suggest adding to Mistake_Patterns.md\n- **Consistent preferences:** User repeatedly mentioned a workflow preference?\n\n> \"This week's session learnings revealed:\n> \n> **Recurring Issues:**\n> - Calendar overload (mentioned 3 times) — Consider blocking focus time\n> \n> **Workflow Preferences:**\n> - Prefer morning for deep work (mentioned 2 times)\n> \n> Should I add these to your pattern files?\"\n\n---\n\n## Output Format\n\nCreate `00-Inbox\u002FYYYY-MM-DD - Weekly Synthesis.md`.\n\n> **Dual-format lookup:** When reading previous weekly synthesis files (e.g., for trend comparison), check both the current format `00-Inbox\u002FYYYY-MM-DD - Weekly Synthesis.md` and the legacy format `00-Inbox\u002FWeekly_Synthesis_YYYY-MM-DD.md`. Prefer the current format if both exist.\n\nTemplate:\n\n```markdown\n# Weekly Synthesis — Week of [Date]\n\n## TL;DR\n\n- **Weekly priorities:** [X] of 3 complete\n- **Tasks:** [X] completed \u002F [Y] planned — [Z]% completion\n- **Meetings:** [N] total\n- **Key wins:** [1-2 bullets]\n- **Carried over:** [1-2 items]\n\n---\n\n## 🎯 Weekly Priorities\n\n### 1. [Priority 1] — ✅ Complete\n\n**Deliverable:** [What was shipped\u002Ffinished]\n**Completed:** [Day]\n**Tasks:** 5 of 5\n\n### 2. [Priority 2] — 🔄 In Progress (60%)\n\n**Done this week:**\n- [Specific accomplishment]\n- [Specific accomplishment]\n\n**Remaining:**\n- [Specific task]\n- [Specific task]\n\n### 3. [Priority 3] — ❌ Not Started\n\n**Why:** [Reason]\n**Recommendation:** [Carry forward \u002F Deprioritize \u002F Defer]\n\n---\n\n## 📊 Task Completion\n\n| Metric | Count |\n|--------|-------|\n| Tasks completed | 14 |\n| Tasks added mid-week | 6 |\n| Tasks carried over | 3 |\n| **Completion rate** | **82%** |\n\n**Observation:** [Any patterns — e.g., lots of scope creep]\n\n---\n\n## 🎯 Quarterly Goals\n\n| Goal | Milestones | This Week | Status |\n|------|------------|-----------|--------|\n| [Goal 1] | X of Y | +Z | [Status] |\n| [Goal 2] | X of Y | — | [Status] |\n| [Goal 3] | X of Y | +Z | [Status] |\n\n**Goals advancing:** [Which ones moved]\n**Goals stalled:** [Which ones need attention]\n\n---\n\n## 📊 Daily Completion Trend\n\n| Day | Planned | Done | Rate |\n|-----|---------|------|------|\n| Mon | 3 | 2 | 67% |\n| Tue | 3 | 3 | 100% |\n| Wed | 3 | 2 | 67% |\n| Thu | 3 | 1 | 33% |\n| Fri | 3 | 2 | 67% |\n\n**Week average:** [X]%\n**Observation:** [Pattern noticed]\n\n---\n\n## 📅 Meetings & People\n\n### Meetings Held\n\n| Date | Topic | Key Outcome |\n|------|-------|-------------|\n| [Day] | [Topic] | [Decision\u002Finsight] |\n\n### New Contacts\n- [Name] at [Company] — [context]\n\n### Action Items from Meetings\n- [ ] [Action] — for [who] — due [when]\n\n---\n\n## 📡 Slack Intelligence\n\n*Channels: [list of channels scanned]*\n*Period: YYYY-MM-DD to YYYY-MM-DD*\n\n### Action Items Captured\n- [ ] [Action] — from #[channel], [date] — [priority\u002Fpillar]\n\n### Project Updates\n- **[Project Name]:** [Key update or context from thread]\n\n### People Signals\n- **[Person]:** [Signal — role change, feedback, preference, etc.]\n\n### Org Intel\n- [Announcement or trend relevant to L&D programs]\n\n---\n\n## 💡 Learnings\n\n### Session Learnings (Auto-Captured)\n- [Learning 1]\n- [Learning 2]\n\n### Patterns Identified\n- **Recurring issue:** [Issue] (appeared X times)\n- **Preference noted:** [Preference]\n\n### Actionable Improvements\n- [ ] [Specific improvement to make]\n\n---\n\n## 📊 Pillar Balance\n\n| Pillar | Tasks Done | Focus |\n|--------|------------|-------|\n| [Pillar 1] | X tasks | Heavy |\n| [Pillar 2] | X tasks | Light |\n| [Pillar 3] | X tasks | None |\n\n**Observation:** [Balance assessment]\n\n---\n\n## ➡️ Next Week\n\n### Suggested Priorities\n\nBased on this week's progress:\n\n1. **[Priority]** — [Why: carries over \u002F goal needs attention \u002F commitment]\n2. **[Priority]** — [Why]\n3. **[Priority]** — [Why]\n\n### Blocked Items Needing Resolution\n\n| Item | Blocked Since | What Would Unblock It |\n|------|---------------|-----------------------|\n| [Item] | [Date] | [Action needed] |\n\n---\n\n## 🏆 Career Evidence (If Career System Enabled)\n\n**Significant accomplishments worth capturing:**\n\n- [Accomplishment] — demonstrates [skill]\n- [Accomplishment] — shows [impact]\n\n> \"Want to save any of these as career evidence?\"\n\n---\n\n*Generated: [timestamp]*\n*Weekly completion: X of 3 priorities*\n*Task completion: X%*\n```\n\n---\n\n## Innovation Concierge: Top 3 This Week\n\nAt the end of the weekly review, surface the top backlog ideas:\n\n1. Call `list_ideas(status=\"active\", min_score=70)` from Improvements MCP\n2. Pick the top 3 ideas by score that haven't been surfaced in the last week review\n3. Include in the output format as a section:\n\n```markdown\n## 🤖 Top 3 Amp Improvement Ideas\n\nYour AI-curated backlog has surfaced these high-impact ideas:\n\n1. **[idea-XXX]** Title (Score: XX)\n   Why now: [Brief evidence or timeliness reason]\n\n2. **[idea-XXX]** Title (Score: XX)\n   Why now: [Brief evidence]\n\n3. **[idea-XXX]** Title (Score: XX)\n   Why now: [Brief evidence]\n\n> Interested? Run `\u002Famp-improve [idea-id]` to workshop any of these.\n> Run `\u002Famp-backlog` to see the full ranked backlog.\n```\n\n**Rules:**\n- Only show ideas with score >= 70 (don't surface low-value noise)\n- Prefer ideas with recent \"Why Now?\" evidence\n- If fewer than 3 qualifying ideas, show however many exist\n- If no qualifying ideas, skip this section entirely\n- This is a gentle nudge, not a sales pitch\n\n---\n\n## Skill Quality Insights\n\nAfter generating the synthesis, call `get_skill_ratings()` from Work MCP (no filter — get all skills).\n\n**If ratings exist for any skills:**\nAdd a section to the review:\n\n```markdown\n## Skill Quality This Week\n\n| Skill | Avg Rating | Trend | Note |\n|-------|-----------|-------|------|\n| [skill] | [avg]\u002F5 | [improving\u002Fstable\u002Fdeclining] | [most recent note] |\n```\n\n**Only surface skills that are declining or below 3.0.** If everything is stable\u002Fgood, skip this section entirely. One line for healthy, only details for problems.\n\n**Then:** Run `\u002Fidentity-snapshot` to update `System\u002Fidentity-model.md` with fresh data from this week.\n\n---\n\n## Vault Hygiene Check (Silent)\n\nRun the vault maintenance checker as part of the weekly review:\n\n```bash\nnode .claude\u002Fhooks\u002Fmaintenance.cjs\n```\n\nIf any issues are found, add a section to the synthesis:\n\n```markdown\n## 🧹 Vault Hygiene\n\n| Check | Count | Action |\n|-------|-------|--------|\n| Stale inbox files (>30d) | [N] | Archive or delete |\n| Broken WikiLinks | [N] | Fix or remove |\n| Orphaned person pages | [N] | Review for relevance |\n| Stale agent memory (>90d) | [N] | Safe to clean up |\n```\n\nIf all counts are zero, skip this section entirely (clean vault, nothing to report).\n\nOnly list the top 3-5 items per category. Don't overwhelm the review with maintenance details.\n\n---\n\n## System Improvement Scan (Proactive)\n\nAfter the main synthesis, scan 7 data sources for improvement opportunities and generate a proposal file. This is the weekly intelligence layer that spots patterns across your entire workflow.\n\n### Data Sources to Scan\n\n**1. Session History (session_store)**\n\nQuery the session store for the past 7 days:\n\n```sql\n-- Most-used tools this week\nSELECT content, source_type FROM search_index\nWHERE search_index MATCH 'tool OR edit OR create OR bash'\nAND source_type = 'turn'\nORDER BY rank LIMIT 20;\n\n-- Sessions with many turns (potential friction)\nSELECT s.id, s.summary, COUNT(t.turn_index) as turns\nFROM sessions s JOIN turns t ON t.session_id = s.id\nWHERE s.created_at >= date('now', '-7 days')\nGROUP BY s.id\nORDER BY turns DESC LIMIT 5;\n\n-- Failed tool calls or retries\nSELECT content FROM search_index\nWHERE search_index MATCH 'error OR failed OR retry OR broken'\nAND source_type = 'turn'\nORDER BY rank LIMIT 10;\n```\n\nLook for:\n- Tools called repeatedly in sequence (automation opportunity)\n- Sessions with 10+ turns for a simple task (friction)\n- Repeated error patterns (reliability issue)\n- Files read\u002Fedited most often (hot paths worth optimizing)\n\n**2. Usage Adoption Gaps**\n\nRead `System\u002Fusage_log.md` and count:\n- Total features available vs. checked off\n- Categories with zero adoption (entire feature areas unused)\n- Features checked recently vs. long ago\n\nCompare against the user's role and pillars. Flag features that would directly help their work but haven't been tried.\n\n**3. Skill Quality Trends**\n\nRead `System\u002FSkill_Ratings\u002Fratings.jsonl` (if it exists):\n- Skills with average rating below 3.0\n- Skills with declining trend (last 3 ratings decreasing)\n- Skills used frequently but never rated (blind spots)\n\nIf no ratings exist, note this as an observation: \"No skill quality data yet. Rating skills after use helps me improve over time.\"\n\n**4. File Structure Growth**\n\nScan key vault directories:\n\n```bash\n# File counts per top-level directory\nfind 00-Inbox -type f -name \"*.md\" | wc -l\nfind 04-Projects -type f -name \"*.md\" | wc -l\nfind 05-Areas -type f -name \"*.md\" | wc -l\n\n# Files not touched in 60+ days\nfind 04-Projects 05-Areas -type f -name \"*.md\" -mtime +60 | head -10\n\n# Recently created files (new growth)\nfind . -type f -name \"*.md\" -mtime -7 -not -path \".\u002Fnode_modules\u002F*\" -not -path \".\u002F.claude\u002F*\" | wc -l\n```\n\nLook for:\n- Inbox growing faster than it's being processed (triage needed)\n- Projects with no recent activity (stale projects)\n- Missing expected files (no daily reviews, empty quarter goals)\n- Naming inconsistencies across similar files\n\n**5. Vault Hygiene**\n\nReuse the output from `maintenance.cjs` already run in this review. Don't re-run it.\n\n**6. Hook & Tool Health**\n\nCheck `.logs\u002Ferror-queue.json` for:\n- Tools that errored 3+ times this week (reliability problem)\n- MCP servers with unacknowledged errors\n- Tools available in MCP config but never called (from session history)\n\nCheck `.claude\u002Fhooks\u002F` for:\n- Hook files that exist but aren't in `.github\u002Fhooks\u002Fhooks.json` (unwired hooks)\n\n**7. Workflow Friction Patterns**\n\nUse Work MCP data:\n\n```\nlist_tasks(include_done=false) — check for:\n  - Tasks in \"started\" status for 5+ days (stuck work)\n  - P0 tasks open more than 3 days (urgency leak)\n\nget_week_priorities() for last 3 weeks — check for:\n  - Same priority appearing 2+ weeks (chronic carryover)\n  - Priorities with no linked tasks (goal without action)\n\nget_quarterly_goals() — check for:\n  - Goals with 0% progress and weeks passing\n  - Goals with no linked weekly priorities\n```\n\n### Synthesis Rules\n\nAfter gathering all observations:\n\n1. **Deduplicate against last proposal.** Read `System\u002FImprovement_Proposals\u002F` for the most recent file. Skip anything you suggested last week that hasn't changed.\n\n2. **Rank by impact.** Use three tiers:\n   - 🔴 **High Impact** — Directly affects daily productivity, saves significant time, or fixes broken workflows. Limit to 1-3 items.\n   - 🟡 **Medium Impact** — Would improve quality of life but isn't urgent. 2-4 items.\n   - 🟢 **Low Impact** — Nice to have, cosmetic, or speculative. 1-3 items.\n\n3. **Be specific.** Every suggestion must include:\n   - What you observed (with data\u002Fcounts)\n   - Why it matters (concrete impact)\n   - What to do about it (actionable fix, not vague advice)\n   - Effort estimate (quick win \u002F medium \u002F deep work)\n   - Files affected\n\n4. **Don't be noisy.** If fewer than 3 total observations are worth flagging, that's fine. \"Nothing new to suggest this week\" is a valid outcome. Don't pad with low-value filler.\n\n5. **Auto-capture high-impact items.** For any 🔴 High Impact suggestion, call `capture_idea()` from the Improvements MCP to add it to the formal backlog (with dedup check).\n\n### Output\n\nWrite the proposal to `System\u002FImprovement_Proposals\u002FYYYY-MM-DD.md`:\n\n```markdown\n# System Improvement Proposal — Week of YYYY-MM-DD\n\n## Summary\n\n[2-3 sentences: X opportunities found across Y categories. Z are quick wins.]\n\n---\n\n## 🔴 High Impact\n\n### 1. [Clear, specific title]\n\n**What I noticed:** [Concrete observation with data. E.g., \"You ran `lookup_person` 14 times this week, but the person context injector hook should be doing this automatically.\"]\n\n**Why it matters:** [Impact. E.g., \"Each manual lookup costs ~5 seconds of context. The hook would make this invisible.\"]\n\n**Suggested fix:** [Specific action. E.g., \"Fix the path in person-context-injector.cjs from 'People\u002F' to '05-Areas\u002FPeople\u002F'. I can do this now.\"]\n\n**Effort:** Quick win\n**Files:** `.claude\u002Fhooks\u002Fperson-context-injector.cjs`\n\n---\n\n## 🟡 Medium Impact\n\n### 2. [Title]\n...\n\n---\n\n## 🟢 Low Impact\n\n### 3. [Title]\n...\n\n---\n\n## 📊 System Health Snapshot\n\n| Metric | This Week | Trend |\n|--------|-----------|-------|\n| Tasks completed | X | ↑\u002F↓\u002F→ |\n| Skills invoked | X unique | |\n| Tool errors | X | |\n| Vault files | X total | |\n| Inbox backlog | X files | |\n\n---\n\n## Skipped\n\n[Brief list of things you considered but decided weren't worth flagging, with one-line reasoning. This shows your work and helps the user understand your judgment.]\n```\n\nOpen the file in Obsidian after writing:\n```bash\nsource scripts\u002Fobsidian-cli.sh && obs_open \"YYYY-MM-DD\"\n```\n\n### Presentation to User\n\nAfter generating the proposal, add a brief section to the week-review output:\n\n```\n## 🔬 System Improvement Proposal\n\nGenerated `System\u002FImprovement_Proposals\u002FYYYY-MM-DD.md` with [N] suggestions:\n- 🔴 [count] high impact: [one-line summary of top item]\n- 🟡 [count] medium impact\n- 🟢 [count] low impact\n\n[Opened in Obsidian for review.]\n```\n\nIf no improvements were found: \"System scan complete. Nothing new to suggest this week.\"\n\n---\n\n## Follow-up Actions\n\nAfter synthesis:\n1. Update Tasks.md with new priorities\n2. Archive completed items\n3. Update project pages with status changes\n4. Offer to run `\u002Fweek-plan` for next week\n\n---\n\n## MCP Dependencies\n\n| Integration | MCP Server | Tools Used |\n|-------------|------------|------------|\n| Work | work-mcp | `list_tasks`, `get_week_progress`, `get_quarterly_goals`, `get_goal_status` |\n| Calendar | calendar-mcp | `calendar_get_events_with_attendees` |\n| Improvements | amp-improvements-mcp | `list_ideas` |\n| Analytics | amp-analytics | `track_event` |\n\n---\n\n## Track Usage (Silent)\n\nUpdate `System\u002Fusage_log.md` to mark weekly review as used.\n\n**Analytics (Silent):**\n\nCall `track_event` with event_name `week_review_completed` and properties:\n- `priorities_completed`: number of priorities completed\n- `priorities_total`: total number of priorities\n- `tasks_completed`: number of tasks completed this week. This only fires if the user has opted into analytics. No action needed if it returns \"analytics_disabled\".\n","## Purpose\n\nCreate a synthesis of the week reviewing activity, progress, and what was accomplished. **Uses concrete metrics, not vague percentages.**\n\n---\n\n## Data Sources\n\n### 1. Task Progress\n- `03-Tasks\u002FTasks.md` — Task completion status\n- `02-Week_Priorities\u002FWeek_Priorities.md` — Weekly priorities\n\n### 2. Project Activity\n- `04-Projects\u002F**\u002F*.md` — Modified this week\n\n### 3. Meetings & People\n- `00-Inbox\u002FMeetings\u002F*.md` — Meeting notes from this week\n- `People\u002F**\u002F*.md` — Person pages updated\n\n### 4. Learnings\n- `06-Resources\u002FLearnings\u002F**\u002F*.md` — Explicit learnings\n- `System\u002FSession_Learnings\u002F*.md` — Auto-captured session learnings\n\n### 5. Daily Reviews\n- `07-Archives\u002FReviews\u002FYYYY-MM-DD - Daily Review.md` — This week's reviews (fall back to `Daily_Review_YYYY-MM-DD.md` for legacy naming)\n\n### 6. Journals (If Enabled)\n- `00-Inbox\u002FJournals\u002FYYYY\u002FMM-Month\u002F` — Morning\u002Fevening journals\n\n### 7. Slack Channels (If Enabled)\n- Configured channels from `System\u002Fuser-profile.yaml` > `slack_intel`\n- Pulled via `gh slack-aggregator review` with weekly date range\n\n---\n\n## Analysis Process\n\n### 1. Weekly Priority Completion (Concrete, Not Percentages)\n\n**Don't say:** \"Goal X went from 40% to 55%\"\n**Do say:** \"You completed 2 of 3 weekly priorities\"\n\n```\nUse: get_week_progress()\n```\n\nFor each weekly priority:\n- **Complete:** ✅ What was the deliverable? When did you finish?\n- **In Progress:** 🔄 What specifically got done? What's left?\n- **Not Started:** ❌ Why? Should it carry forward?\n\n**Surface concrete accomplishments:**\n\n> \"**This week's priorities:**\n> \n> 1. ✅ **Ship pricing page** — Complete (pushed to prod Wednesday)\n>    - Deliverable: New pricing page live at \u002Fpricing\n>    - Tasks completed: 5 of 5\n> \n> 2. 🔄 **Write Q1 strategy doc** — 60% complete\n>    - Done: Outline, competitive analysis, recommendations\n>    - Remaining: Executive summary, financial projections\n>    - 2 tasks left\n> \n> 3. ❌ **Customer interviews** — Not started\n>    - Reason: Calendar was too stacked\n>    - Recommendation: Carry to next week with protected time\"\n\n### 1.5 Semantic Goal-to-Work Mapping (if QMD available)\n\n**Check if semantic search is available** by looking for `qmd` in PATH.\n\nIf available, enhance the weekly priority review with meaning-based analysis:\n\n1. **Auto-detect goal contributions:** For each completed task this week, search:\n   ```\n   qmd query \"task title\u002Fdescription\" --limit 3\n   ```\n   against quarterly goals. Catch tasks that advanced goals without explicit links.\n   - Example: \"Built customer health dashboard\" semantically matches goal \"Improve NPS tracking\" — different words, same work.\n\n2. **Cross-priority connections:** Search for work that bridges multiple priorities:\n   ```\n   qmd query \"priority 1 description\" --limit 5\n   ```\n   Surface tasks that contributed to more than one priority.\n\n3. **Thematic patterns:** Search for recurring themes across the week's work:\n   ```\n   qmd query \"common theme from meetings\u002Ftasks\" --limit 5\n   ```\n   Detect patterns like \"most of your work this week clustered around customer retention\" even when tasks used different terminology.\n\n**Integration:** Merge findings into the Quarterly Goals table — add a \"Hidden contributions\" row for semantically-detected but not explicitly-linked work. Only show genuinely new connections, not things already captured by keyword matching.\n\n**If QMD unavailable:** Skip silently. Task completion stats still work fine.\n\n### 2. Task Completion Stats (Concrete Numbers)\n\nScan `03-Tasks\u002FTasks.md` for completion timestamps from this week:\n- Count tasks completed (look for `✅ YYYY-MM-DD` in date range)\n- Count tasks added mid-week\n- Count tasks carried over\n\n**Surface:**\n\n> \"**Tasks this week:**\n> - Completed: 14 tasks\n> - Added mid-week: 6 tasks (scope creep?)\n> - Carried over: 3 tasks\n> \n> **Completion rate:** 82% (14 of 17 planned)\"\n\n### 3. Quarterly Goals Progress (Concrete Milestones)\n\n**Don't use fake percentages.** Use milestone counts and specific accomplishments.\n\n```\nUse: get_quarterly_goals()\nUse: get_goal_status(goal_id) for each goal\n```\n\nFor each goal:\n- Milestones completed this week\n- Total milestones done vs. total\n- Weeks since last milestone\n- Specific accomplishments that moved the goal\n\n> \"**Quarterly Goals Progress:**\n> \n> | Goal | Milestones | This Week | Status |\n> |------|------------|-----------|--------|\n> | Launch v2.0 | 3 of 5 | +1 (Pricing page shipped) | On track |\n> | Improve NPS | 1 of 4 | No change | ⚠️ Stalled (3 weeks) |\n> | Team Capacity | 2 of 3 | No change | On track |\n> \n> **Goal 1** advanced because you completed Priority 1.\n> **Goal 2** needs attention — no linked work completed this week.\"\n\n### 4. Daily Completion Rate Trend (NEW)\n\nIf daily reviews exist, calculate completion trends:\n\n> \"**Daily plan completion this week:**\n> \n> | Day | Planned | Done | Rate |\n> |-----|---------|------|------|\n> | Mon | 3 | 2 | 67% |\n> | Tue | 3 | 3 | 100% |\n> | Wed | 3 | 2 | 67% |\n> | Thu | 3 | 1 | 33% |\n> | Fri | 3 | 2 | 67% |\n> \n> **Week average:** 67%\n> **Pattern:** Thursday was rough (too many meetings?)\"\n\n### 5. Meeting Analysis\n\nReview meeting notes from the week:\n- Meetings held\n- Key decisions\n- Action items created\n- Follow-ups that might have slipped\n\n### 5.5 Commitment Health Analysis (NEW)\n\nIf ScreenPipe and Commitment Detection are available, show aggregate stats:\n\n```\nUse: get_commitment_stats(\n    start_date=\"YYYY-MM-DD\",  # Monday of this week\n    end_date=\"YYYY-MM-DD\"     # Today\n)\n```\n\n**Surface to user:**\n\n> \"📊 **Commitment Health This Week**\n>\n> **Detected across apps:** 12 potential commitments\n> **Already had tasks:** 7 (58%)\n> **Created from prompts:** 3\n> **Dismissed as handled:** 2\n>\n> **Apps with most uncaptured asks:**\n> 1. Slack - 5 items\n> 2. Email - 4 items\n> 3. Notion - 3 items\n>\n> **People who asked most of you:**\n> 1. Sarah Chen - 4 asks\n> 2. Product team - 3 asks\n>\n> 💡 *Consider: Check Slack more frequently for asks, or run `\u002Fcommitment-scan` mid-week*\"\n\n**If no commitment data:**\nSkip this section silently (user may not have ScreenPipe or commitment detection enabled).\n\n### 5.8 Email Communication Stats (if Gmail connected)\n\nCheck `System\u002Fintegrations\u002Fconfig.yaml` for `google-workspace.enabled: true`.\n\nIf enabled and Google Workspace MCP is healthy:\n- **Emails sent this week** — count of sent messages in the review period\n- **Average response time** — how quickly you replied to incoming emails\n- **Threads still open** — conversations with no resolution (back-and-forth still active)\n- **Follow-up detection** — emails waiting > 48h for a reply from you or from others\n\nSurface in the review:\n\n> \"**Email this week:**\n>\n> | Metric | Value |\n> |--------|-------|\n> | Emails sent | 47 |\n> | Avg response time | 3.2 hours |\n> | Open threads | 12 |\n> | Awaiting your reply (> 48h) | 3 |\n>\n> **Observation:** You have 3 emails waiting for replies longer than 48 hours. Consider clearing those early next week.\"\n\nIf unhealthy or not enabled: skip this section silently.\n\n### 5.9 Slack Intelligence (if configured)\n\nCheck `System\u002Fuser-profile.yaml` > `slack_intel.enabled`.\n\nIf enabled and `gh slack-aggregator` is installed, run the `\u002Fslack-intel` skill workflow:\n\n1. Pull each configured channel using `gh slack-aggregator review --channel \u003Curl> --from \u003Cmonday> --to \u003Ctoday>`\n2. Analyze exported markdown against active projects, people, and pillars\n3. Categorize findings: 🎯 Action Items, 📋 Project Context, 👤 People Signals, 📊 Org Intel\n4. Present findings to the user for confirmation and routing\n5. Route confirmed items to tasks, project notes, person pages\n6. Save summary to `00-Inbox\u002FSlack_Intel\u002FWeek_of_YYYY-MM-DD.md`\n7. Clean up temporary export files\n\n**Full workflow details:** See `.claude\u002Fskills\u002Fslack-intel-custom\u002FSKILL.md`\n\nSurface in the review:\n\n> \"📡 **Slack Intelligence**\n>\n> **Channels scanned:** 3\n> **Items found:** X action items, X project updates, X people signals\n>\n> [Categorized findings presented for user confirmation]\"\n\nAction items from Slack Intel feed into the \"Next Week\" suggestions section.\n\nIf not enabled or aggregator not installed: skip this section silently.\n\n### 6. Learning Compilation & Pattern Detection\n\nReview `System\u002FSession_Learnings\u002F` files from this week:\n\n**Pattern Detection:**\n- **Recurring issues:** Same mistake 2+ times? Suggest adding to Mistake_Patterns.md\n- **Consistent preferences:** User repeatedly mentioned a workflow preference?\n\n> \"This week's session learnings revealed:\n> \n> **Recurring Issues:**\n> - Calendar overload (mentioned 3 times) — Consider blocking focus time\n> \n> **Workflow Preferences:**\n> - Prefer morning for deep work (mentioned 2 times)\n> \n> Should I add these to your pattern files?\"\n\n---\n\n## Output Format\n\nCreate `00-Inbox\u002FYYYY-MM-DD - Weekly Synthesis.md`.\n\n> **Dual-format lookup:** When reading previous weekly synthesis files (e.g., for trend comparison), check both the current format `00-Inbox\u002FYYYY-MM-DD - Weekly Synthesis.md` and the legacy format `00-Inbox\u002FWeekly_Synthesis_YYYY-MM-DD.md`. Prefer the current format if both exist.\n\nTemplate:\n\n```markdown\n# Weekly Synthesis — Week of [Date]\n\n## TL;DR\n\n- **Weekly priorities:** [X] of 3 complete\n- **Tasks:** [X] completed \u002F [Y] planned — [Z]% completion\n- **Meetings:** [N] total\n- **Key wins:** [1-2 bullets]\n- **Carried over:** [1-2 items]\n\n---\n\n## 🎯 Weekly Priorities\n\n### 1. [Priority 1] — ✅ Complete\n\n**Deliverable:** [What was shipped\u002Ffinished]\n**Completed:** [Day]\n**Tasks:** 5 of 5\n\n### 2. [Priority 2] — 🔄 In Progress (60%)\n\n**Done this week:**\n- [Specific accomplishment]\n- [Specific accomplishment]\n\n**Remaining:**\n- [Specific task]\n- [Specific task]\n\n### 3. [Priority 3] — ❌ Not Started\n\n**Why:** [Reason]\n**Recommendation:** [Carry forward \u002F Deprioritize \u002F Defer]\n\n---\n\n## 📊 Task Completion\n\n| Metric | Count |\n|--------|-------|\n| Tasks completed | 14 |\n| Tasks added mid-week | 6 |\n| Tasks carried over | 3 |\n| **Completion rate** | **82%** |\n\n**Observation:** [Any patterns — e.g., lots of scope creep]\n\n---\n\n## 🎯 Quarterly Goals\n\n| Goal | Milestones | This Week | Status |\n|------|------------|-----------|--------|\n| [Goal 1] | X of Y | +Z | [Status] |\n| [Goal 2] | X of Y | — | [Status] |\n| [Goal 3] | X of Y | +Z | [Status] |\n\n**Goals advancing:** [Which ones moved]\n**Goals stalled:** [Which ones need attention]\n\n---\n\n## 📊 Daily Completion Trend\n\n| Day | Planned | Done | Rate |\n|-----|---------|------|------|\n| Mon | 3 | 2 | 67% |\n| Tue | 3 | 3 | 100% |\n| Wed | 3 | 2 | 67% |\n| Thu | 3 | 1 | 33% |\n| Fri | 3 | 2 | 67% |\n\n**Week average:** [X]%\n**Observation:** [Pattern noticed]\n\n---\n\n## 📅 Meetings & People\n\n### Meetings Held\n\n| Date | Topic | Key Outcome |\n|------|-------|-------------|\n| [Day] | [Topic] | [Decision\u002Finsight] |\n\n### New Contacts\n- [Name] at [Company] — [context]\n\n### Action Items from Meetings\n- [ ] [Action] — for [who] — due [when]\n\n---\n\n## 📡 Slack Intelligence\n\n*Channels: [list of channels scanned]*\n*Period: YYYY-MM-DD to YYYY-MM-DD*\n\n### Action Items Captured\n- [ ] [Action] — from #[channel], [date] — [priority\u002Fpillar]\n\n### Project Updates\n- **[Project Name]:** [Key update or context from thread]\n\n### People Signals\n- **[Person]:** [Signal — role change, feedback, preference, etc.]\n\n### Org Intel\n- [Announcement or trend relevant to L&D programs]\n\n---\n\n## 💡 Learnings\n\n### Session Learnings (Auto-Captured)\n- [Learning 1]\n- [Learning 2]\n\n### Patterns Identified\n- **Recurring issue:** [Issue] (appeared X times)\n- **Preference noted:** [Preference]\n\n### Actionable Improvements\n- [ ] [Specific improvement to make]\n\n---\n\n## 📊 Pillar Balance\n\n| Pillar | Tasks Done | Focus |\n|--------|------------|-------|\n| [Pillar 1] | X tasks | Heavy |\n| [Pillar 2] | X tasks | Light |\n| [Pillar 3] | X tasks | None |\n\n**Observation:** [Balance assessment]\n\n---\n\n## ➡️ Next Week\n\n### Suggested Priorities\n\nBased on this week's progress:\n\n1. **[Priority]** — [Why: carries over \u002F goal needs attention \u002F commitment]\n2. **[Priority]** — [Why]\n3. **[Priority]** — [Why]\n\n### Blocked Items Needing Resolution\n\n| Item | Blocked Since | What Would Unblock It |\n|------|---------------|-----------------------|\n| [Item] | [Date] | [Action needed] |\n\n---\n\n## 🏆 Career Evidence (If Career System Enabled)\n\n**Significant accomplishments worth capturing:**\n\n- [Accomplishment] — demonstrates [skill]\n- [Accomplishment] — shows [impact]\n\n> \"Want to save any of these as career evidence?\"\n\n---\n\n*Generated: [timestamp]*\n*Weekly completion: X of 3 priorities*\n*Task completion: X%*\n```\n\n---\n\n## Innovation Concierge: Top 3 This Week\n\nAt the end of the weekly review, surface the top backlog ideas:\n\n1. Call `list_ideas(status=\"active\", min_score=70)` from Improvements MCP\n2. Pick the top 3 ideas by score that haven't been surfaced in the last week review\n3. Include in the output format as a section:\n\n```markdown\n## 🤖 Top 3 Amp Improvement Ideas\n\nYour AI-curated backlog has surfaced these high-impact ideas:\n\n1. **[idea-XXX]** Title (Score: XX)\n   Why now: [Brief evidence or timeliness reason]\n\n2. **[idea-XXX]** Title (Score: XX)\n   Why now: [Brief evidence]\n\n3. **[idea-XXX]** Title (Score: XX)\n   Why now: [Brief evidence]\n\n> Interested? Run `\u002Famp-improve [idea-id]` to workshop any of these.\n> Run `\u002Famp-backlog` to see the full ranked backlog.\n```\n\n**Rules:**\n- Only show ideas with score >= 70 (don't surface low-value noise)\n- Prefer ideas with recent \"Why Now?\" evidence\n- If fewer than 3 qualifying ideas, show however many exist\n- If no qualifying ideas, skip this section entirely\n- This is a gentle nudge, not a sales pitch\n\n---\n\n## Skill Quality Insights\n\nAfter generating the synthesis, call `get_skill_ratings()` from Work MCP (no filter — get all skills).\n\n**If ratings exist for any skills:**\nAdd a section to the review:\n\n```markdown\n## Skill Quality This Week\n\n| Skill | Avg Rating | Trend | Note |\n|-------|-----------|-------|------|\n| [skill] | [avg]\u002F5 | [improving\u002Fstable\u002Fdeclining] | [most recent note] |\n```\n\n**Only surface skills that are declining or below 3.0.** If everything is stable\u002Fgood, skip this section entirely. One line for healthy, only details for problems.\n\n**Then:** Run `\u002Fidentity-snapshot` to update `System\u002Fidentity-model.md` with fresh data from this week.\n\n---\n\n## Vault Hygiene Check (Silent)\n\nRun the vault maintenance checker as part of the weekly review:\n\n```bash\nnode .claude\u002Fhooks\u002Fmaintenance.cjs\n```\n\nIf any issues are found, add a section to the synthesis:\n\n```markdown\n## 🧹 Vault Hygiene\n\n| Check | Count | Action |\n|-------|-------|--------|\n| Stale inbox files (>30d) | [N] | Archive or delete |\n| Broken WikiLinks | [N] | Fix or remove |\n| Orphaned person pages | [N] | Review for relevance |\n| Stale agent memory (>90d) | [N] | Safe to clean up |\n```\n\nIf all counts are zero, skip this section entirely (clean vault, nothing to report).\n\nOnly list the top 3-5 items per category. Don't overwhelm the review with maintenance details.\n\n---\n\n## System Improvement Scan (Proactive)\n\nAfter the main synthesis, scan 7 data sources for improvement opportunities and generate a proposal file. This is the weekly intelligence layer that spots patterns across your entire workflow.\n\n### Data Sources to Scan\n\n**1. Session History (session_store)**\n\nQuery the session store for the past 7 days:\n\n```sql\n-- Most-used tools this week\nSELECT content, source_type FROM search_index\nWHERE search_index MATCH 'tool OR edit OR create OR bash'\nAND source_type = 'turn'\nORDER BY rank LIMIT 20;\n\n-- Sessions with many turns (potential friction)\nSELECT s.id, s.summary, COUNT(t.turn_index) as turns\nFROM sessions s JOIN turns t ON t.session_id = s.id\nWHERE s.created_at >= date('now', '-7 days')\nGROUP BY s.id\nORDER BY turns DESC LIMIT 5;\n\n-- Failed tool calls or retries\nSELECT content FROM search_index\nWHERE search_index MATCH 'error OR failed OR retry OR broken'\nAND source_type = 'turn'\nORDER BY rank LIMIT 10;\n```\n\nLook for:\n- Tools called repeatedly in sequence (automation opportunity)\n- Sessions with 10+ turns for a simple task (friction)\n- Repeated error patterns (reliability issue)\n- Files read\u002Fedited most often (hot paths worth optimizing)\n\n**2. Usage Adoption Gaps**\n\nRead `System\u002Fusage_log.md` and count:\n- Total features available vs. checked off\n- Categories with zero adoption (entire feature areas unused)\n- Features checked recently vs. long ago\n\nCompare against the user's role and pillars. Flag features that would directly help their work but haven't been tried.\n\n**3. Skill Quality Trends**\n\nRead `System\u002FSkill_Ratings\u002Fratings.jsonl` (if it exists):\n- Skills with average rating below 3.0\n- Skills with declining trend (last 3 ratings decreasing)\n- Skills used frequently but never rated (blind spots)\n\nIf no ratings exist, note this as an observation: \"No skill quality data yet. Rating skills after use helps me improve over time.\"\n\n**4. File Structure Growth**\n\nScan key vault directories:\n\n```bash\n# File counts per top-level directory\nfind 00-Inbox -type f -name \"*.md\" | wc -l\nfind 04-Projects -type f -name \"*.md\" | wc -l\nfind 05-Areas -type f -name \"*.md\" | wc -l\n\n# Files not touched in 60+ days\nfind 04-Projects 05-Areas -type f -name \"*.md\" -mtime +60 | head -10\n\n# Recently created files (new growth)\nfind . -type f -name \"*.md\" -mtime -7 -not -path \".\u002Fnode_modules\u002F*\" -not -path \".\u002F.claude\u002F*\" | wc -l\n```\n\nLook for:\n- Inbox growing faster than it's being processed (triage needed)\n- Projects with no recent activity (stale projects)\n- Missing expected files (no daily reviews, empty quarter goals)\n- Naming inconsistencies across similar files\n\n**5. Vault Hygiene**\n\nReuse the output from `maintenance.cjs` already run in this review. Don't re-run it.\n\n**6. Hook & Tool Health**\n\nCheck `.logs\u002Ferror-queue.json` for:\n- Tools that errored 3+ times this week (reliability problem)\n- MCP servers with unacknowledged errors\n- Tools available in MCP config but never called (from session history)\n\nCheck `.claude\u002Fhooks\u002F` for:\n- Hook files that exist but aren't in `.github\u002Fhooks\u002Fhooks.json` (unwired hooks)\n\n**7. Workflow Friction Patterns**\n\nUse Work MCP data:\n\n```\nlist_tasks(include_done=false) — check for:\n  - Tasks in \"started\" status for 5+ days (stuck work)\n  - P0 tasks open more than 3 days (urgency leak)\n\nget_week_priorities() for last 3 weeks — check for:\n  - Same priority appearing 2+ weeks (chronic carryover)\n  - Priorities with no linked tasks (goal without action)\n\nget_quarterly_goals() — check for:\n  - Goals with 0% progress and weeks passing\n  - Goals with no linked weekly priorities\n```\n\n### Synthesis Rules\n\nAfter gathering all observations:\n\n1. **Deduplicate against last proposal.** Read `System\u002FImprovement_Proposals\u002F` for the most recent file. Skip anything you suggested last week that hasn't changed.\n\n2. **Rank by impact.** Use three tiers:\n   - 🔴 **High Impact** — Directly affects daily productivity, saves significant time, or fixes broken workflows. Limit to 1-3 items.\n   - 🟡 **Medium Impact** — Would improve quality of life but isn't urgent. 2-4 items.\n   - 🟢 **Low Impact** — Nice to have, cosmetic, or speculative. 1-3 items.\n\n3. **Be specific.** Every suggestion must include:\n   - What you observed (with data\u002Fcounts)\n   - Why it matters (concrete impact)\n   - What to do about it (actionable fix, not vague advice)\n   - Effort estimate (quick win \u002F medium \u002F deep work)\n   - Files affected\n\n4. **Don't be noisy.** If fewer than 3 total observations are worth flagging, that's fine. \"Nothing new to suggest this week\" is a valid outcome. Don't pad with low-value filler.\n\n5. **Auto-capture high-impact items.** For any 🔴 High Impact suggestion, call `capture_idea()` from the Improvements MCP to add it to the formal backlog (with dedup check).\n\n### Output\n\nWrite the proposal to `System\u002FImprovement_Proposals\u002FYYYY-MM-DD.md`:\n\n```markdown\n# System Improvement Proposal — Week of YYYY-MM-DD\n\n## Summary\n\n[2-3 sentences: X opportunities found across Y categories. Z are quick wins.]\n\n---\n\n## 🔴 High Impact\n\n### 1. [Clear, specific title]\n\n**What I noticed:** [Concrete observation with data. E.g., \"You ran `lookup_person` 14 times this week, but the person context injector hook should be doing this automatically.\"]\n\n**Why it matters:** [Impact. E.g., \"Each manual lookup costs ~5 seconds of context. The hook would make this invisible.\"]\n\n**Suggested fix:** [Specific action. E.g., \"Fix the path in person-context-injector.cjs from 'People\u002F' to '05-Areas\u002FPeople\u002F'. I can do this now.\"]\n\n**Effort:** Quick win\n**Files:** `.claude\u002Fhooks\u002Fperson-context-injector.cjs`\n\n---\n\n## 🟡 Medium Impact\n\n### 2. [Title]\n...\n\n---\n\n## 🟢 Low Impact\n\n### 3. [Title]\n...\n\n---\n\n## 📊 System Health Snapshot\n\n| Metric | This Week | Trend |\n|--------|-----------|-------|\n| Tasks completed | X | ↑\u002F↓\u002F→ |\n| Skills invoked | X unique | |\n| Tool errors | X | |\n| Vault files | X total | |\n| Inbox backlog | X files | |\n\n---\n\n## Skipped\n\n[Brief list of things you considered but decided weren't worth flagging, with one-line reasoning. This shows your work and helps the user understand your judgment.]\n```\n\nOpen the file in Obsidian after writing:\n```bash\nsource scripts\u002Fobsidian-cli.sh && obs_open \"YYYY-MM-DD\"\n```\n\n### Presentation to User\n\nAfter generating the proposal, add a brief section to the week-review output:\n\n```\n## 🔬 System Improvement Proposal\n\nGenerated `System\u002FImprovement_Proposals\u002FYYYY-MM-DD.md` with [N] suggestions:\n- 🔴 [count] high impact: [one-line summary of top item]\n- 🟡 [count] medium impact\n- 🟢 [count] low impact\n\n[Opened in Obsidian for review.]\n```\n\nIf no improvements were found: \"System scan complete. Nothing new to suggest this week.\"\n\n---\n\n## Follow-up Actions\n\nAfter synthesis:\n1. Update Tasks.md with new priorities\n2. Archive completed items\n3. Update project pages with status changes\n4. Offer to run `\u002Fweek-plan` for next week\n\n---\n\n## MCP Dependencies\n\n| Integration | MCP Server | Tools Used |\n|-------------|------------|------------|\n| Work | work-mcp | `list_tasks`, `get_week_progress`, `get_quarterly_goals`, `get_goal_status` |\n| Calendar | calendar-mcp | `calendar_get_events_with_attendees` |\n| Improvements | amp-improvements-mcp | `list_ideas` |\n| Analytics | amp-analytics | `track_event` |\n\n---\n\n## Track Usage (Silent)\n\nUpdate `System\u002Fusage_log.md` to mark weekly review as used.\n\n**Analytics (Silent):**\n\nCall `track_event` with event_name `week_review_completed` and properties:\n- `priorities_completed`: number of priorities completed\n- `priorities_total`: total number of priorities\n- `tasks_completed`: number of tasks completed this week. This only fires if the user has opted into analytics. No action needed if it returns \"analytics_disabled\".\n",{"name":5,"context":24,"description":7},"fork","main","6a447b9ff35eed7bda9e12399bcd373dd31e0a53",".claude\u002Fskills\u002Fweek-review","https:\u002F\u002Fgithub.com\u002Fhemory\u002Famp","https:\u002F\u002Fgithub.com\u002Fhemory","https:\u002F\u002Fwww.heminthehat.com\u002F","The People Developer | L&D Program Manager | Software Engineer","2026-06-09 00:20:04",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add hemory\u002Famp --skill week-review",{"tool":39,"command":40},"bunx","bunx skills add hemory\u002Famp --skill week-review",{"tool":42,"command":43},"pnpm","pnpm dlx skills add hemory\u002Famp --skill week-review","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fhemory\u002Famp\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fweek-review\u002FSKILL.md",true,[],[48,55,62,69,76,83],{"skillNo":49,"name":50,"nameZh":6,"description":51,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":52,"categories":53,"keywords":54,"publishedAt":19},"01M76W5GGTP0YHZQQJ4G2ZPSZE","zoom-setup","Connect Zoom to Amp for meeting recordings, scheduling, and transcript context","https:\u002F\u002Fgithub.com\u002Fhemory\u002Famp\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fzoom-setup\u002FSKILL.md",[],[],{"skillNo":56,"name":57,"nameZh":6,"description":58,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":59,"categories":60,"keywords":61,"publishedAt":19},"01M76W5CWR8MABE0XPZVRXFMDK","xray","Understand what just happened under the hood - learn AI by seeing it in action","https:\u002F\u002Fgithub.com\u002Fhemory\u002Famp\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fxray\u002FSKILL.md",[],[],{"skillNo":63,"name":64,"nameZh":6,"description":65,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":66,"categories":67,"keywords":68,"publishedAt":19},"01M76W54EY8D8Y55DEPYPJ9YBV","week-plan","Set weekly priorities and plan the week ahead with intelligent suggestions based on goals, calendar shape, and task effort.","https:\u002F\u002Fgithub.com\u002Fhemory\u002Famp\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fweek-plan\u002FSKILL.md",[],[],{"skillNo":70,"name":71,"nameZh":6,"description":72,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"publishedAt":19},"01M76W527EJ43YXBN0CE9SRAYT","triage","Strategically route orphaned files and extract scattered tasks","https:\u002F\u002Fgithub.com\u002Fhemory\u002Famp\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Ftriage\u002FSKILL.md",[],[],{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":19},"01M76W4ZYRDVHZ3SS8W5WZ53Q8","trello-setup","Connect Trello to Amp for visual Kanban task sync","https:\u002F\u002Fgithub.com\u002Fhemory\u002Famp\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Ftrello-setup\u002FSKILL.md",[],[],{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"publishedAt":19},"01M76W4WXPFTAQT7KM8E65QVBE","todoist-setup","Connect Todoist to Amp for two-way task sync","https:\u002F\u002Fgithub.com\u002Fhemory\u002Famp\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Ftodoist-setup\u002FSKILL.md",[],[]]