ReviewSAFE
A complete starter kit for an Obsidian + Claude Code personal knowledge management system.
Overview
A complete starter kit for an Obsidian + Claude Code personal knowledge management system.
855f65f39078OBSERVED · 2026-10-08What it tells the agent
The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.
--- name: review description: Smart review router. Detects context (morning, Sunday, end of month) and launches the appropriate review workflow. Use anytime for the right review at the right time. allowed-tools: Read, Write, Edit, Glob, Grep, Bash, TaskCreate, TaskUpdate, TaskList, TaskGet model: sonnet user-invocable: true --- # Review Skill Smart router that detects context and launches the appropriate review workflow. ## Usage ``` /review # Auto-detect the right review based on time/context /review daily # Force daily review /review weekly # Force weekly review /review monthly # Force monthly review ``` Or simply: "Help me review" — and the right workflow starts. ## Auto-Detection Logic When invoked without arguments, detect context using these rules: ### 1. Check the Time of Day ```bash HOUR=$(date +%H) ``` - **Before noon (< 12):** Morning routine — delegate to `/daily` morning workflow - **After 5 PM (>= 17):** Evening shutdown — delegate to `/daily` evening workflow - **Midday (12-17):** Midday check-in — delegate to `/daily` midday workflow ### 2. Check the Day of Week ```bash DAY_OF_WEEK=$(date +%u) # 1=Monday, 7=Sunday ``` - **Sunday (7) or Monday (1):** Weekly review — delegate to `/weekly` - Override time-of-day detection - Ask: "Ready for your weekly review?" before proceeding ### 3. Check the Day of Month ```bash DAY_OF_MONTH=$(date +%d) DAYS_IN_MONTH=$(date -v+1m -v1d -v-1d +%d 2>/dev/null || date -d "$(date +%Y-%m-01) +1 month -1 day" +%d) ``` - **Last 3 days of month (DAY_OF_MONTH >= DAYS_IN_MONTH - 2):** Monthly review — delegate to `/monthly` - Override both time-of-day and day-of-week detection - Ask: "End of month — ready for your monthly review?" before proceeding - **First day of month (DAY_OF_MONTH == 1):** Also suggest monthly review - "It's the first of the month. Want to do your monthly review for last month?" ### 4. Check Staleness Before routing, check for overdue reviews: ```bash # Read weekly review file for last date WEEKLY_REVIEW="Goals/3. Weekly Review.md" # If last weekly review > 7 days ago, suggest weekly regardless of day ``` - **Weekly review overdue (>7 days):** Suggest weekly review - "Your last weekly review was N days ago. Want to catch up?" - If user says no, fall through to time-of-day detection ## Routing Behavior After detecting context: 1. Tell the user what was detected: "It's Sunday evening — launching your weekly review." 2. Delegate to the appropriate skill's workflow 3. The delegated skill handles everything from there ### Delegation This skill does NOT duplicate the logic of `/daily`, `/weekly`, or `/monthly`. It: 1. Detects context 2. Informs the user 3. Follows the instructions from the target skill's SKILL.md ### Explicit Override If the user specifies a type (`/review weekly`), skip auto-detection entirely and go directly to that review type. ## Output on Detection ```markdown ### Review Router **Time:** 7:15 AM (Morning) **Day:** Sunday **Month day:** 15th **Detected:** Weekly review (Sunday override) **Last weekly review:** 3 days ago (not overdue) Launching weekly review... ``` ## Edge Cases - **Multiple triggers** (e.g., last Sunday of month): Monthly takes priority over weekly - **No daily note exists**: Create one first, then continue with review - **User says "no" to suggestion**: Fall through to next detection level - **Explicit argument overrides everything**: `/review monthly` runs monthly review even on a Tuesday morning ## Integration Works with: - `/daily` — Morning, midday, and evening routines - `/weekly` — Full weekly review process - `/monthly` — Monthly review and planning - Session init hook — Staleness data already calculated
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | NA |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (0)
No findings outside the package's declared scope.
Gates applied: no_behavioural_pass.
855f65f39078full audit observations/trust-audit/skill/ballred__review.json · Report an issue / request a re-scanAudit history
Every audit this skill has had.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-08 | 855f65f39078 | SAFE | B | 89 | first audit |
Questions
What does the Review skill do?
A complete starter kit for an Obsidian + Claude Code personal knowledge management system.
Is Review safe to install?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean skill reads B.
What can Review access on my machine?
The audit observed no filesystem, network or shell use at all in its source.
How current is this page?
The grade is for one exact copy of the source (855f65f39078), read on 2026-10-08. The repository is watched, and a new audit runs when it changes — this is the first audit.