agentleFS
nikolajhh2008-svgnikolajhh2008-svg/brainwarden/skills/brain-ask/SKILL.md

brain-ask

Skill2 starsChanged 4 months agoMITView on GitHub

Answer questions from the Brain with cited notes. Use when the user says "what does my brain know about …", "ask my brain", "what did I decide about …", "who is …", "what's the status of …", or when any question is about their own life, people, projects or past decisions.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. mshtawythug/second-brain★ 16same name

    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.

  • In this repository5 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 1 is distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    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. Very specific:

    Orient, then search wide: Read <vault>/index.md first — the root signpost names the mode, the vault language, the folder map and the search …

    line 22
  2. Very specific:

    Decisions are special: if the question touches a past decision, quote the decision record (40-decisions/) including its reasoning — that's …

    line 34
  3. Very specific:

    Offer the next step: when a gap is researchable: "want me to research that and work it in?" (→ skill brain-research) — or capture the open …

    line 49
  4. Distinctive:

    Every claim in the answer must be traceable to a note or clearly labeled as NOT from the vault.

    line 54

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Brain ask (the librarian)

Chats evaporate — the brain doesn't. This skill answers questions from the human's OWN notes, shows where every claim comes from, and is honest about what the brain does not know.

Conventions: <vault> = the vault path from the Brain vault: line in your global rules — or, if there is no such line, the folder this session was started in, when that folder holds a CLAUDE.md naming a vault mode. A shared company vault has no global line on purpose (its skills travel inside it), so falling back to ~/Brain there would write into somebody's private vault. Only when neither exists does ~/Brain apply. python3 = your working python command (on most Windows machines py -3; the global rules name it).

Steps

  1. Orient, then search wide: Read <vault>/index.md first — the root signpost names the mode, the vault language, the folder map and the search tool, and unless this session started inside the vault you have none of that. (No index.md? Read <vault>/CLAUDE.md instead.) Then skim <vault>/Home.md (the living dashboard often frames the answer) and run python3 <vault>/.tools/search.py <terms> with 2–3 term variants (synonyms, the person's name, the project slug). Read ONLY the hits — never the whole vault.
  2. Answer from the notes, in the vault language, citing every source note as a [[link]] so they can click into it in Obsidian. The notes' content outranks your general knowledge — if the vault and your training data disagree about the human's life, the vault wins.
  3. Decisions are special: if the question touches a past decision, quote the decision record (40-decisions/) including its reasoning — that's exactly what the append-only log is for. A record marked Superseded by … is history: answer from the one that superseded it and name both.
  4. Two hits that contradict each other? Do not pick one. Give both, with their dates, say which is newer and where each comes from — a moved date usually leaves the old one standing somewhere. Then offer to capture the correction. Answering confidently from one of two conflicting notes is the worst thing this skill can do: it sounds certain and sends the person to last month's appointment. A vault that contradicts itself is a finding, not a malfunction.
  5. Say what's missing, plainly: "your brain has nothing on X" is a correct and useful answer — name the closest note you did find so they can judge the gap. Never fill gaps with invented vault content.
  6. Offer the next step when a gap is researchable: "want me to research that and work it in?" (→ skill brain-research) — or capture the open question to the inbox if they answer it themselves on the spot.

Rules

  • Every claim in the answer must be traceable to a note or clearly labeled as NOT from the vault.
  • In company mode (the vault CLAUDE.md names it) this is absolute: answer only from notes, cite the file path for every statement, and say "I don't know" rather than reason your way to a plausible number. Notes marked status: draft are proposals, not company truth — label them.
  • Short questions get short answers — cite, don't lecture.
  • Works from ANY session, whatever project is open.

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.