agentleFS
paslavskyipaslavskyi/llm-wiki/.claude/skills/kb-orient/SKILL.md

kb-orient

Skill1 starsChanged 32 days agoMITView on GitHub

Use at the start of every session in this knowledge repo to learn where we are and what's next, before any other work. Reads STATE and the index map, never the whole repo.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 1 is distinctive, and 6 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.

  • 6

    Very specific

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

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

    Read kb.config.yml. Note mode and language.

    line 14
  2. Very specific:

    Read STATE.md — current focus + next step (intent ONLY.

    line 17
  3. Very specific:

    Read index/MAP.md — domains and note counts.

    line 18
  4. Very specific:

    If index/health.md exists, read it for open questions/risks.

    line 20
  5. Very specific:

    Run npm run plan -- show — the operational plan (в розробці / черга / backlog + recent done).

    line 21
  6. Very specific:

    Never read the whole knowledge/ tree here.

    line 34
  7. Distinctive:

    Do not modify any file in this skill.

    line 35

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

kb-orient — session bootstrap

Goal: in seconds, know the current phase, what's done, what's next, and open questions — without loading the whole knowledge base.

Steps

  1. The SessionStart hook already put a [kb] session snapshot in context (focus, plan, attention counts) — start from it; the steps below expand it.
  2. Read kb.config.yml. Note mode and language. Respond in language.
    • debug: report what you read and the full picture.
    • autonomous: give only a short human summary of where things stand.
  3. Read STATE.md — current focus + next step (intent ONLY; it does not hold counts/progress).
  4. Read index/MAP.md — domains and note counts. Do NOT open individual notes yet. This is the AUTHORITATIVE source for counts/progress; if STATE.md ever disagrees, MAP/health win.
  5. If index/health.md exists, read it for open questions/risks.
  6. Run npm run plan -- show — the operational plan (в розробці / черга / backlog + recent done). This is the delivery half of "where we are"; STATE holds only strategic intent. Goals/timeline: index/roadmap.index.md if needed.
  7. Produce an orientation summary:
    • Current phase and the single recommended next step.
    • Plan snapshot: what's in work, what's next in the queue.
    • Per-domain note counts (from MAP).
    • Any open questions/risks you saw.
    • Offer the live dashboard for visuals (kb-dashboard → http://localhost:4321).
  8. Hand off: if the next step is knowledge capture, suggest running kb-elicit or proceed with kb-capture for a specific insight.

Rules

  • Never read the whole knowledge/ tree here. Orientation is index-only.
  • Do not modify any file in this skill.

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.