agentleFS
QuarktexQuarktex/qcobrain/.claude/skills/recall/SKILL.md

recall

Skill4 starsChanged 4 months agoMITView on GitHub

Answer any question about you, your business, the market, or a past decision from your co-brain — with exact verbatim citations back to `sources/`. Flow - grep `INDEX.md` → read the candidate notes → re-verify each `src` quote against its source → answer with citations. Refuses to fabricate: if it isn't sourced, it says so plainly and offers to `/capture` the gap. Triggers - any factual/business question, "recall", "what does my brain say", "what did I decide about X", "who is Y", "what do I know about Z".

151 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use addskill, apply-firmware 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.
  • ryanportfolioSkill

    addskill

    Create, import, update, or install repo or personal skills: runtime ownership, resources, discovery checks.

    Also in 6 projects that use this

  • ryanportfolioSkill

    apply-firmware

    Bring current folder (plain folder, existing repo, old or partial firmware copy) up to full Harness Firmware in place. Mirror external repo → adopt-repo.

    Also in 6 projects that use this

  • +10 more from ryanportfolio

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 2 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 138 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

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

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

    Read the shortlisted wiki/ note bodies.

    line 20
  2. Distinctive:

    If two notes contradict, surface BOTH and say it's unresolved — don't paper over it.

    line 33
  3. Very specific:

    If no sourced note answers the question, say so plainly: "Nothing sourced in the wiki covers this." Then offer leads — the closest wiki/ …

    line 36
  4. Distinctive:

    No verifiable source → no asserted fact.

    line 42

“Common” is measured: kept by at least 6 in 10 of the 138 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

Inherits references/kernel.md (the provenance contract). Reads references/wiki-spec.md for schema.

The whole point of this skill: answers that are faster AND more accurate than you are from memory — with exact sources. An answer without a citation is a bug.

Execution

Step 1 — Find candidate notes (no body reads yet)

  1. grep INDEX.md — the NOTES table, the ALIASES section, and the SUPERSEDED redirects — for keywords from the question. Aliases catch phrasing variance; superseded redirects stop you citing retired facts.
  2. Shortlist the matching note ids. Follow any superseded_by to the current note.

Step 2 — Read and re-verify

  1. Read the shortlisted wiki/ note bodies.
  2. For each src block you're about to rely on, re-verify the citation: resolve source via sources/_sources.md to a path, then confirm the quote still appears verbatim in that file. If the source file's hash drifted, re-check; if the quote no longer resolves, the note is stale — flag it, don't silently trust it (and suggest /wiki-doctor).

Step 3 — Answer

  1. Answer in your voice (see references/voice.md / CLAUDE.md). Direct, concise, lead with the answer.
  2. Every factual claim is followed by its citation. Format: (source: <source-key>, "<short verbatim quote>", captured <date>) — and name the wiki/ note id when useful.
  3. Distinguish confidence honestly: "the source states…" (stated) vs "I inferred from…" (inferred) vs "reportedly, per {{party}}…" (reported). For a reported note, keep the attribution visible — a third-party claim is never restated as your own fact.
  4. If two notes contradict, surface BOTH and say it's unresolved — don't paper over it.

Step 4 — When the brain doesn't know

  1. If no sourced note answers the question, say so plainly: "Nothing sourced in the wiki covers this." Then offer leads — the closest wiki/ notes, any related inbox/ drafts (flagged as unsourced) — and an offer to run /capture to fill the gap. Never fabricate an answer or a citation to fill the silence.

Refusal rule

No verifiable source → no asserted fact. Surface the gap and the nearest leads instead. A confident, unsourced answer is the single worst failure mode this co-brain can produce.

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.