Audit log

Skill torlyai/Schengen-master/skills/audit-log

Session-local action history for the vstack toolkit. Shows what skills have been run in the current Claude Code session, what decisions were made, and what's still outstanding. Useful when the user comes back from a coffee break and asks "where were we" or needs to summarise for a family member. Strictly in-memory; no cross-session persistence (use /learn for that). Use when the user says "what have we done", "summary", "where are we", or "list our decisions". (Schengen-master skills)From its SKILL.md

Install

npx -y skills add torlyai/Schengen-master --skill audit-log

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

3 stars3 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

7.3 KB, as published. Nobody here has run it

/audit-log

What this skill does

You are the Schengen-master Meta Utility (audit log). You produce an in-session summary: which skills the user has run, what decisions were made, what's still outstanding.

This is session-local only. It doesn't persist (use /learn for cross-session). It's the "where are we?" answer.

Apply ETHOS principle #4 ("Surface decisions, not just actions") — the user cares about what was decided, not which commands were typed.

When to use this skill

User loses track in a complex session (e.g. family of 4 with sponsor)
User wants to share a summary with their spouse / sponsor
User comes back from a break ("OK, where were we?")
User wants to confirm they've covered everything before /audit-application
Mid-session pause to take stock

What to surface (and what not to)

Surface	Hide
Skills run + outcomes	Internal Claude reasoning steps
Decisions made (visa type, TLS centre, etc.)	Verbatim conversation transcripts
Documents drafted (cover letter, sponsor letter, etc.)	Full document contents
Outstanding action items	Lengthy file diffs
Critical-path deadlines	Tooling implementation details

The format is a digest — what a busy applicant needs to glance at and re-orient.

Procedure

Scan the current session for skill invocations and outcomes
Group by category:
- Trip planning decisions
- Document collection status
- Financial situation
- Timeline + critical-path
- Outstanding actions
Render the digest in the output template

Output template

SESSION AUDIT LOG
Session started: {{TIME}}
Skills run: {{COUNT}}

═════════════════════════════════════════════════════════════════════
TRIP PLAN
═════════════════════════════════════════════════════════════════════

{{IF_PLANNED:
  Destination: {{COUNTRY/CITIES}}
  Dates: {{TRAVEL_DATES}}
  Type: {{VISA_TYPE}}
  Purpose: {{PURPOSE}}
  Group: {{COMPOSITION}}
}}
{{IF_NOT_PLANNED: Trip not yet locked. Run /plan-trip.}}

═════════════════════════════════════════════════════════════════════
DOCUMENTS
═════════════════════════════════════════════════════════════════════

✅ Done:
  • {{ITEM_1}}
  • {{ITEM_2}}

⏳ In progress:
  • {{ITEM_3}}
  • {{ITEM_4}}

❌ Outstanding:
  • {{ITEM_5}}
  • {{ITEM_6}}

═════════════════════════════════════════════════════════════════════
DECISIONS MADE
═════════════════════════════════════════════════════════════════════

  • TLS centre: {{CENTRE}}
  • Visa type: {{TYPE}}
  • Number of entries: {{ENTRIES}}
  • Sponsor: {{NAME_OR_NONE}}
  • {{OTHER_DECISIONS}}

═════════════════════════════════════════════════════════════════════
TIMELINE
═════════════════════════════════════════════════════════════════════

Today: {{DATE}}
TLS appointment booked: {{DATE_OR_TBD}}
Trip date: {{DATE}}
Critical-path deadline: {{DATE_OF_BOTTLENECK}}

═════════════════════════════════════════════════════════════════════
SUGGESTED NEXT STEP
═════════════════════════════════════════════════════════════════════

{{NEXT_SKILL_RECOMMENDATION}}

═════════════════════════════════════════════════════════════════════
WANT TO PERSIST THIS?
═════════════════════════════════════════════════════════════════════

Run /learn to save this state across sessions.
Otherwise it's session-local only.

Routing rules

Situation	Suggest next
Many items outstanding, deadline tight	`/timeline-planner`
Documents 80%+ done	`/audit-application`
Trip not yet shaped	`/plan-trip`
User wants to share with family	Copy the audit log; send via Slack/email
User wants this to persist	`/learn`
Session getting long, user confused	Run `/audit-log`, then take a break

Common pitfalls

Pitfall	Why it hurts	Fix
Treating audit log as the application	It's a session digest, not the truth	Trust the actual documents, not the log
Forgetting that audit log is session-local	Disappears when session ends	Use `/learn` for cross-session
Showing too much detail	Overwhelms	Keep digest format
Surfacing internal Claude reasoning	Confuses user	Show outcomes, not deliberation
Audit log replacing real preparation	Audit log doesn't make decisions	It's a viewer, not an editor

Authoritative sources

Session memory only

Notes for maintainers

The audit log is a "viewer" skill — it shouldn't modify any state, just summarize what's been done.
For very long sessions (50+ messages), aim for a 1-screen digest. Anything longer loses utility.
Format consistency matters: use the same emoji/symbol set as other skill outputs (✅ ⏳ ❌).
For users who don't speak comfortable English, the digest is especially useful as a translation-friendly summary.
This skill is a good place to suggest /learn ("want to save this?") since users running /audit-log are usually at a natural pause point.
For multi-applicant sessions, consider per-applicant breakdowns instead of one massive log.