agentleFS
Sign inStart free
MeshJSMeshJS/evolving-claude-skills/.claude/skills/wrap-up/SKILL.md

wrap-up

Skill4 starsChanged 9 months agoApache-2.0View on GitHub

End-of-session automation. Creates a journey, consolidates learning notes, evolves skills, cross-pollinates shared patterns.

We split this skill into the rules it gives an agent

13 rules. 0 are common, 8 are distinctive, and 5 are very specific to this project.

  • 0

    Common

    Most of the 57 edited versions we read keep these.

  • 8

    Distinctive

    This author's own choices. Worth a look.

  • 5

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    CRITICAL: Before anything else, run git commands to capture ALL changes.

    line 32
  2. Distinctive:

    All modified files and what changed

    line 41
  3. Distinctive:

    All new files created

    line 42
  4. Distinctive:

    Search ALL skill files for inline learning notes left during the session:

    line 66
  5. Distinctive:

    After consolidation, the section reads naturally with the learning absorbed:

    line 89
  6. Very specific:

    Read all skills from .claude/skills/*/SKILL.md and compare against journey learnings.

    line 99
  7. Very specific:

    Update the updated date in frontmatter

    line 110
  8. Distinctive:

    Check if any learnings from the session apply across multiple skills.

    line 116
  9. Very specific:

    Read .claude/skills/_patterns.md (create it if it doesn't exist yet)

    line 118
  10. Very specific:

    If a learning is cross-cutting (applies to multiple skills or domains), add it to _patterns.md

    line 119
  11. Very specific:

    If a pattern in _patterns.md is outdated based on session work, update it

    line 120
  12. Distinctive:

    If something looks stale, remove it.

    line 136
  13. Distinctive:

    Add a section to the journey documenting skill changes:

    line 142

“Common” is measured: kept by at least 6 in 10 of the 57 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

Session Wrap-Up

The single learning engine for the skill system. Mirrors how learning works in nature:

Experience (work) → Awareness (learning notes) → Consolidation (wrap-up)

Usage

/wrap-up {session title}

Arguments

  • $0 - Session title describing the main work done (e.g., "API Auth Refactor")

Step 1: Gather All File Changes

CRITICAL: Before anything else, run git commands to capture ALL changes. In long sessions, memory alone will miss things.

git status
git diff HEAD
git ls-files --others --exclude-standard

From the output, build a comprehensive list of:

  • All modified files and what changed
  • All new files created
  • All deleted files
  • Nature of changes (bug fix, feature, refactor, etc.)

DO NOT proceed to Step 2 until git output is reviewed.

Step 2: Create Journey

Using file changes from Step 1 AND conversation context, create a journey at .claude/journeys/{date}-{slug}.md.

Include:

  • Summary: What was accomplished and why it matters
  • What Was Done: Numbered list of major items
  • Key Learnings: Insights that prevent future mistakes
  • Files Changed: Table of files and changes
  • Patterns Discovered: Code examples of reusable patterns
  • Decisions Made: Table of decisions and rationale

Step 3: Process Learning Notes

Search ALL skill files for inline learning notes left during the session:

# Find all learning notes across skills
grep -r "<!-- LEARNING" .claude/skills/

For each learning note found:

  1. Read the note and its surrounding context in the SKILL.md
  2. Integrate the learning into the skill's proper text — rewrite the relevant section to include the new knowledge naturally
  3. Remove the raw note — the <!-- LEARNING ... --> comment gets replaced by the integrated content
  4. Update the updated date in the skill's frontmatter

Learning Note Format (for reference)

During work, learning notes are left inline:

### Some Section
Existing instruction text...

<!-- LEARNING 2026-02-08: Discovered that X actually needs Y because Z -->

After consolidation, the section reads naturally with the learning absorbed:

### Some Section
Existing instruction text. Note: X needs Y because Z.

Step 4: Evolve Skills from Journey

Read all skills from .claude/skills/*/SKILL.md and compare against journey learnings.

For each skill, check:

  • Does the journey contain patterns the skill should document?
  • Were there edge cases or gotchas the skill should warn about?
  • Are there new verification checklist items?
  • Did we discover something that contradicts current skill instructions?

For each skill that needs updates:

  1. Read current SKILL.md
  2. Apply changes based on learnings
  3. Update the updated date in frontmatter

Step 5: Cross-Pollinate Shared Patterns

Check if any learnings from the session apply across multiple skills.

  1. Read .claude/skills/_patterns.md (create it if it doesn't exist yet)
  2. If a learning is cross-cutting (applies to multiple skills or domains), add it to _patterns.md
  3. If a pattern in _patterns.md is outdated based on session work, update it

Examples of cross-cutting learnings:

  • A new error handling pattern discovered while building an API endpoint → add to shared patterns
  • A database query optimization found during feature work → add to shared patterns
  • A validation pattern discovered in form handling → add to shared patterns

Step 6: Prune Stale Content

Review skills touched during the session. Look for:

  • Sections documenting patterns for deprecated/removed features
  • Duplicate content that now lives in _patterns.md
  • Instructions that conflict with current codebase reality

If something looks stale, remove it. Keep skills lean — context window space is valuable.

Step 7: Update Journey with Evolutions

Add a section to the journey documenting skill changes:

## Skills Evolved

| Skill | Changes |
|-------|---------|
| skill-name | Added error handling pattern |
| _patterns | Updated validation rules |

Skill Evolution Guidelines

Session ContentAction
New component/module patternAdd to relevant skill's patterns section
Bug fix with root causeAdd to gotchas/checklist in relevant skill
New feature implementedUpdate feature documentation in relevant skill
Cross-cutting patternUpdate _patterns.md
New convention discoveredUpdate _patterns.md

Example Output

Session: API Auth Refactor

1. Gathered file changes (8 modified, 2 created)
2. Created journey: .claude/journeys/2026-02-08-api-auth-refactor.md
3. Processed 2 learning notes:
   - auth-flow: integrated token refresh edge case
   - api-client: integrated retry pattern
4. Evolved 2 skills from journey learnings:
   - auth-flow: added session expiry handling
   - api-client: added graceful degradation
5. Cross-pollinated: added error retry pattern to _patterns.md
6. Pruned: removed outdated middleware example from api-client
7. Updated journey with evolution summary

Download now (ZIP)

  • Other versions62 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 62

    1. rohitg00/pro-workflow★ 2.9ksame name
    2. anthropics/launch-your-agent★ 1ksame name
    3. ianlkl11234s/mini-taiwan-pulse★ 531same name
    4. aspi6246/Claude-Code-Skills-for-Academics★ 153same name
    5. PaulRBerg/agent-skills★ 104same name
    6. KerberosClaw/kc_ai_skills★ 79same name
    7. jonathanmalkin/jules★ 65same name
    8. simoncorry/foundry★ 39same name
    9. alxyrgin/health-os★ 38same name
    10. mshadmanrahman/claude-for-designers★ 32same name

    52 more not listed

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • Used alongsideProjects with it also use sync, orchestration 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.
    • MicrosoftSkill

      sync

      Sync the current session branch with its upstream branch, or publish the current session branch to a remote. Use when the user asks to sync a branch, pull latest changes, rebase onto upstream, push current branch, publish branch, or set upstream.

      Also in 4 projects that use this

    • austintgriffithSkill

      orchestration

      How an AI agent plans, builds, and deploys a complete Ethereum dApp. The three-phase build system for Scaffold-ETH 2 projects. Use when building a full application on Ethereum — from contracts to frontend to production deployment on IPFS.

      Also in 3 projects that use this

    • SentrySkill

      spec

      Produce a verified spec — problem, desired outcome, and acceptance criteria — before building. Use when given a GitHub/Linear issue link, when starting from a fuzzy in-conversation idea, or asked to "figure out what to do" / "scope this" / "what should I build" — whenever the *what* isn't yet pinned down. Grills the ambiguities, checks the premise against the code, and produces acceptance criteria you sign off on — then hands to design-first.

      Also in 4 projects that use this

    • GitHubSkill

      autoresearch

      Autonomous iterative experimentation loop for any programming task. Guides the user through defining goals, measurable metrics, and scope constraints, then runs an autonomous loop of code changes, testing, measuring, and keeping/discarding results. Inspired by Karpathy''s autoresearch. USE FOR: autonomous improvement, iterative optimization, experiment loop, auto research, performance tuning, automated experimentation, hill climbing, try things automatically, optimize code, run experiments, autonomous coding loop. DO NOT USE FOR: one-shot tasks, simple bug fixes, code review, or tasks without a measurable metric.

      Also in 3 projects that use this

    • seikaikyoSkill

      interview-me

      需求不明時的意圖萃取訪談：一次一題、每題附上自己的猜測、聽出「真正想要 vs 覺得應該要」，直到能預測使用者反應（約 95% 信心）才動工。適用：需求缺少對象 / 動機 / 成功標準 / 約束，或使用者點名「訪談我」「先確認一下」「我們確定嗎」。明確自足的指示、純資訊查詢、機械性操作不適用。

      Also in 3 projects that use this

    • AnthropicSkill

      debug

      Structured debugging session — reproduce, isolate, diagnose, and fix. Trigger with an error message or stack trace, "this works in staging but not prod", "something broke after the deploy", or when behavior diverges from expected and the cause isn't obvious.

      Also in 4 projects that use this

  • In this repository2 other files it gives its agents.

Does it work?

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

Reviews can't be read right now.

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.

Check your setup, free