agentleFS
WolfgangrushWolfgangrush/ai-brain-uk/.claude/skills/retrospective/SKILL.md

retrospective

Skill1 starsChanged 3 months agoMITView on GitHub

End-of-session save discipline for UK AI Brain users. Runs 2-pass leak-check (Rule 6: no internal-shorthand, no internal-paths, no personal-build references). Saves session learning so the firm gets smarter with every use.

73 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use project-stage-detect, team-level and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • CoralGameSkill

    project-stage-detect

    Automatically analyze project state, detect stage, identify gaps, and recommend next steps based on existing artifacts.

    Also in 17 projects that use this

  • CoralGameSkill

    team-level

    Orchestrate level design team: level-designer + narrative-director + world-builder + art-director + systems-designer + qa-tester for complete area/level creation.

    Also in 12 projects that use this

  • lilangMaxSkill

    dev-story

    Read a story file and implement it. Loads the full context (story, GDD requirement, ADR guidelines, control manifest), routes to the right programmer agent for the system and engine, implements the code and test, and confirms each acceptance criterion. The core implementation skill — run after /story-readiness, before /code-review and /story-done.

    Also in 12 projects that use this

  • lilangMaxSkill

    create-stories

    Break a single epic into implementable story files. Reads the epic, its GDD, governing ADRs, and control manifest. Each story embeds its GDD requirement TR-ID, ADR guidance, acceptance criteria, story type, and test evidence path. Run after /create-epics for each epic.

    Also in 11 projects that use this

  • +1 more from CoralGame
  • +7 more from lilangMax

The file

We split this skill into the rules it gives an agent

8 rules. 0 are common, 8 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 62 edited versions we read keep these.

  • 8

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Never reference user-home dot-directories or absolute paths from a developer's machine

    line 75
  2. Distinctive:

    Never use internal entity codes, matter shorthand, or non-jurisdiction-context identifiers

    line 76
  3. Distinctive:

    Never reference external project-architecture names not tied to this firm

    line 77
  4. Distinctive:

    Never reference the publisher's personal builds, personal-vs-business markers, or personal handles

    line 78
  5. Distinctive:

    Does NOT read or write to MemPalace

    line 82
  6. Distinctive:

    Does NOT access personal diary, KG, or self-assessment data

    line 83
  7. Distinctive:

    Does NOT touch any other firm's directory

    line 84
  8. Distinctive:

    Does NOT auto-save without passing leak-check

    line 85

“Common” is measured: kept by at least 6 in 10 of the 62 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

/retrospective — End-of-Session Save Discipline (UK)

Saves what the UK AI Brain learned this session. Runs at session end. The firm accumulates knowledge across sessions without leaking personal context.

2-Pass Leak-Check (Rule 6 — execute BEFORE any save)

Pass 1 — Internal-shorthand codes + internal paths

# Detect any 2-5 letter all-caps token (potential internal matter shorthand)
grep -rniE '\b[A-Z]{2,5}\b' --include="*.md" --include="*.json" --include="*.txt" . 2>/dev/null \
  | grep -vE '\b(USA|UK|EU|UAE|HC|SC|API|JSON|MIT|MCP|CSV|PDF|XML|HTML|HTTP|URL|ABN|TFN|GST|VAT|CEO|CFO|CIO|MD|QC|SC|JD|LLM|LLB|BSc|MBA|GDP|GDPR|CCPA|DPDP|AI|ML|NLP|CLI|IDE|SDK|README|CI|CD|PR|RFC|ETA|EOD|TBD|FYI|NB|AM|PM)\b'
# Detect references to user-home dot-directories (potential machine-internal paths)
grep -rniE '~/\.[a-z]+/' --include="*.md" --include="*.json" . 2>/dev/null

Expected: zero hits. Any hit = STOP. Do not save. Flag for manual review.

Pass 2 — External architecture references + personal-vs-business markers

# Detect references to external project-architecture names (extend this regex with your firm's internal terms)
grep -rniE 'YOUR_INTERNAL_ARCHITECTURE_PATTERN_HERE' --include="*.md" --include="*.json" . 2>/dev/null
# Detect personal-build references (extend with your firm's specific personal-vs-business markers if any)
grep -rniE 'YOUR_PERSONAL_VS_BUSINESS_MARKER_HERE' --include="*.md" --include="*.json" . 2>/dev/null

Expected: zero hits. Any hit = STOP.

Workflow

Step 1: Collect session summary

Review the current conversation for:

  • Which specialists were invoked (Receptionist, Matter Manager, Citation Clerk, Court Registrar, Drafting Assistant, Compliance Officer, Deadline Tracker)
  • What legal domains were touched (civil procedure under CPR, crime, family, chancery, King's Bench, SRA compliance, BSB regulation, UK GDPR, REUL, commercial, property, immigration, employment)
  • What courts or jurisdictions were referenced (UK Supreme Court, EWCA, High Court, Court of Session (Scotland), High Court of Justiciary (Scotland), County Court, Magistrates' Court, tribunals)
  • Any errors or blockers encountered

Step 2: Run 2-pass leak-check (above)

Step 3: If passes leak-check, write session summary

Save to .aibrain-uk/sessions/ with timestamp. Format:

# Session — YYYY-MM-DD HH:MM
- Specialists used: [list]
- Domains: [list]
- Courts: [list]
- Key outcomes: [brief]
- Leak-check: PASS (2-pass, Rule 6)

Step 4: Display save confirmation

🧠 UK AI Brain — session saved.
   Leak-check: ✅ PASS (Pass 1: 0 shorthand, 0 paths | Pass 2: 0 personal-build)
   Next session starts smarter.

Output Format

## /retrospective — UK AI Brain · Session Save

**Session:** YYYY-MM-DD HH:MM
**Specialists active:** [N]
**Domains touched:** [list]
**Leak-check:** ✅ PASS / ❌ BLOCKED

[If PASS: "Session learning saved. Firm knowledge base updated."]
[If BLOCKED: "Leak detected. Session NOT saved. Review flagged content above."]

Anti-Pollution Rules (DO NOT BREAK)

  • Never reference user-home dot-directories or absolute paths from a developer's machine
  • Never use internal entity codes, matter shorthand, or non-jurisdiction-context identifiers
  • Never reference external project-architecture names not tied to this firm
  • Never reference the publisher's personal builds, personal-vs-business markers, or personal handles
  • The publisher credit line in README.md is the sole exception — it is public-facing attribution, not a leak

What this skill does NOT do

  • Does NOT read or write to MemPalace
  • Does NOT access personal diary, KG, or self-assessment data
  • Does NOT touch any other firm's directory
  • Does NOT auto-save without passing leak-check

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.