agentleFS
huntiehuntie/skills/reflect/SKILL.md

reflect

Skill0 starsChanged 21 days agoMITView on GitHub

File what a session taught into vault/, a skill, or a script, applying the vault's write bar. Use at the end of a task when the user corrected the session, a decision changed how a repo's setup or workflow works, or a gotcha cost real time and is not recorded in the code. Also on "reflect" or "remember this". Without one of those triggers, do not use it.

172 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.
  • ianclearySkill

    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 14 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 24 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 19 projects that use this

  • ianclearySkill

    principle-explain-the-number

    Apply 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 26 projects that use this

  • jaydubya818Skill

    automate-me

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

    Also in 39 projects that use this

  • jaydubya818Skill

    poteto-mode

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

    Also in 27 projects that use this

  • +6 more from jaydubya818

The file

We split this skill into the rules it gives an agent

1 rule. 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 163 edited versions we read keep these.

  • 0

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read vault/README.md for the layout and the bar, then vault/index.md.

    line 13

“Common” is measured: kept by at least 6 in 10 of the 163 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

This is the write path into vault/ for decisions and lessons. Filing nothing is the expected outcome of most runs, and the summary says so. One-offs are not learnings.

  1. Read vault/README.md for the layout and the bar, then vault/index.md. Read any existing note or skill a candidate could extend before writing.
  2. Scan the conversation for candidates: corrections received, decisions that changed a workflow, gotchas that cost real time, dead ends and the path that worked when it generalizes, friction in a skill or script this workspace owns, a skill that should have fired and did not, repeated manual steps a script could absorb.
  3. Scrutinize each candidate, in order, and drop it at the first failure:
    • State it as one sentence naming the future task it changes. No such sentence, no note.
    • Durable: still true in six months once paths, SHAs, and tool versions have changed. A platitude fails too; so does a fact that drifts.
    • Drop anything the code, git history, or a few minutes of reading shows on its own. The README's worked example is the calibration.
    • Drop anything already captured. If an existing note or skill covers it partly, the result is an edit that makes it fire, wording or placement, not a new one.
  4. Decide each survivor's route without writing yet, in this order: a script or check that enforces it; the description of a skill that existed but did not trigger; the skill in skills/ it is about; the plan it belongs to; a new skills/principle-<name>/SKILL.md when the rule holds across repos and no skill can hold it, in the shape of the existing principle skills; or, for a note about one repo, the line that repo's own agent instructions would carry.
  5. Present every proposal as one tick list and wait. Use the harness's multiple-choice question with multiple selection where it exists, in batches of four; otherwise a numbered list in prose. Each item is one line: the route and target, then the change or the drafted note, with the recommended items marked. A repo note carries its own destination choice beside it: upstream, as a PR to the repo's AGENTS.md or CLAUDE.md on a task branch or the strongest check the repo would accept; or the overlay under memory/<repo>/, for what the repo cannot hold. Dropped candidates are not options; they go in the summary with their reason.
  6. Apply only what was ticked, then commit each vault write on its own. While touching a repo's overlay, graduate or delete any note that now fails the test: cited twice means it becomes a skill or a script, and one the repo would take goes upstream. A lesson stays under fifteen lines, and a note a survivor supersedes is deleted.

End with a short summary: what was applied and where, one line each; what was offered and declined; and every candidate dropped before the list, with the reason.

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.