agentleFS
0xSoftBoi0xSoftBoi/suwappubot/.claude/skills/reflect/SKILL.md

reflect

Skill4 starsChanged 3 months agoApache-2.0View on GitHub

End-of-task reflection - mine THIS session for corrections and friction, then make surgical bounded edits to lessons.md (and only lessons.md). Use at the end of a task, after the user corrected you, or when invoked as /reflect.

183 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use poteto-help, setup-pstack 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.
  • shiwenbin1617Skill

    poteto-help

    Guides users through pstack setup, /poteto-mode, and picking the skill, playbook, or principle for a task. Type /poteto-help with a question.

    Also in 16 projects that use this

  • bnemaSkill

    setup-pstack

    Configure which Pi models pstack uses per role. Detects available provider/model specs and writes pstack's user configuration. Use for /setup-pstack, "configure pstack models", or changing pstack model choices.

    Also in 27 projects that use this

  • eryzerzSkill

    make-bot-ui

    Use when building a custom UI (page, dashboard, buttons) that should wake a Grok Bot over a webhook, when the user must provide a webhook sender key, or when exposing that UI on Tailscale.

    Also in 20 projects that use this

  • jaydubya818Skill

    automate-me

    Use when the user wants working preferences captured in a reusable mode skill.

    Also in 43 projects that use this

  • jaydubya818Skill

    poteto-mode

    Use when non-trivial work should follow the smallest rigorous Jstack workflow.

    Also in 30 projects that use this

  • shiwenbin1617Skill

    principle-explain-the-number

    Apply before you measure, and before you trust, report, or act on a number you measured: a speedup, a regression, a throughput, a latency, or an eval result. Find what limits it, and rule out that it measured something other than the work you think.

    Also in 29 projects that use this

  • +6 more from jaydubya818

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 3 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 173 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Mine the session you are in: Look for, in priority order:

    line 14
  2. Distinctive:

    If the session contains none of these, say so and stop — do not invent lessons.

    line 20
  3. Very specific:

    Check for an existing lesson first: Read .claude/harness/lessons.md.

    line 22
  4. Very specific:

    If the lesson you're touching has been edited 3+ times, it is not being absorbed as prose — flag it in your reply as a candidate for …

    line 33
  5. Very specific:

    Must PASS. Then commit just that file: HUSKY=0 git add .claude/harness/lessons.md && HUSKY=0 git commit -m "harness(reflect): <lesson …

    line 41
  6. Distinctive:

    Do not push unless the user asked or you are already pushing other work.

    line 43

“Common” is measured: kept by at least 6 in 10 of the 173 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

/reflect — capture what this session taught us

The capture half of the self-improving harness (docs/harness/self-improving.md). /reflect writes lessons; /evolve changes behavior artifacts (CLAUDE.md, hooks, skills, agents). Do not edit anything except .claude/harness/lessons.md here.

Procedure

  1. Mine the session you are in. Look for, in priority order:
    • Explicit user corrections ("no, do X", "I told you", repeated instructions).
    • Tool-call failures you retried more than once (wrong flags, wrong paths, wrong service, permission denials).
    • Anything you discovered the hard way about this repo/deploy that isn't written down (a URL, a quirk, a command that works). If the session contains none of these, say so and stop — do not invent lessons.
  2. Check for an existing lesson first. Read .claude/harness/lessons.md.
    • Same lesson exists → do nothing (or sharpen its wording in place).
    • Related lesson exists → merge into it rather than adding a sibling.
    • New → add under ## Rules (behavior) or ## Facts (environment memory), using the file's format. Max 3 lines per lesson.
  3. Respect the budget. Cap is 25 lessons. At cap: merge or evict the least-valuable lesson in the same edit. Never push the file over cap.
  4. Recidivism check. Run git log --oneline -- .claude/harness/lessons.md | head -20. If the lesson you're touching has been edited 3+ times, it is not being absorbed as prose — flag it in your reply as a candidate for /evolve to promote into an enforcing artifact (hook, skill, or CLAUDE.md rule).
  5. Gate, then commit.
    python3 scripts/harness/harness_lint.py
    Must PASS. Then commit just that file: HUSKY=0 git add .claude/harness/lessons.md && HUSKY=0 git commit -m "harness(reflect): <lesson title>" Do not push unless the user asked or you are already pushing other work.

Output

Reply with: lessons added/merged/evicted (titles only), any recidivism flags, and lint result. Keep it under 6 lines.

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.