agentleFS
Sign inStart free
alphazedealphazede/bearing-lite/skills/scribe/SKILL.md

scribe

Skill3 starsChanged 3 months agoApache-2.0View on GitHub

Transcribe Lifecycle decisions, authority events, ledger snapshots, and profile selections as an event side lane. Use for Scribe. Do not use to activate authority, invent choices, or save profiles unless the owner asks.

13 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 2 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 11 edited versions we read keep these.

  • 2

    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. Very specific:

    Follow ../bearing-lite/references/owner-stops.md for typed owner-stop records: preserve question and round-trip IDs, record actual …

    line 34
  2. Very specific:

    Do not activate authority or invent unresolved choices.

    line 38
  3. Very specific:

    Do not write Lifecycle selections into profiles.json unless the owner asks to save a reusable profile.

    line 43
  4. Distinctive:

    After a genuine context reset/inactive session, resume from the last durable receipt and only missing events under existing resume …

    line 46
  5. Distinctive:

    Never implement, self-assure, select models, or start another invocation without authority.

    line 55

“Common” is measured: kept by at least 6 in 10 of the 11 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

Scribe

Event side lane. Transcribes; cannot activate authority. Loop: consume source-backed events → transcribe → self-check coverage → repair own transcription → recheck → return. Use only a profile-enabled assigned event scope.

Inputs and match

  • Inputs: owner decisions, authority events, combined alignment/scope snapshots, profile selections, configuration digest, journey.json.
  • Match: Scribe is enabled and assigned a visible event scope. Honor the frozen roles.scribe.cadence: event, phase, or lifecycle. One event assignment consumes notifications in one persistent Lifecycle session; never relaunch per event or phase. All cadences send only new events or batches to that session. Missing event transport is a typed gap; keep untranscribed records visible.
  • Non-match: planning dispatch, specialist engineering, execution, or a disabled/ omitted Scribe. Orchestrator transcribes by default when this role is not selected.

Algorithm

  1. Append owner decisions, authority events, ledger snapshots, profile selections, configuration digest, and observed landings (landings: merges and issue closes) to journey.json history. Each history entry carries the source record's id; coverage matches on it. Follow ../bearing-lite/references/owner-stops.md for typed owner-stop records: preserve question and round-trip IDs, record actual asked/answered UTC times and explicit fully blocked intervals, and never infer missing times. Automatic resolutions and owner steering are not invented questions.
  2. Do not activate authority or invent unresolved choices. Finish with node <plugin root>/hooks/scribe-coverage.cjs <journey.json>; return READY only on its PASS. Repair your own transcription errors and rerun the coverage check in this assignment; missing source records are not fabricated. For an open stream, wait/wake on events without polling or timer loops.
  3. Do not write Lifecycle selections into profiles.json unless the owner asks to save a reusable profile.

After a genuine context reset/inactive session, resume from the last durable receipt and only missing events under existing resume authority; preserve scope, route and spent limits. Never reload or retranscribe the whole Lifecycle history.

Return and recovery

Return READY or OWNER_DECISION_REQUIRED with verdict, candidate_ref, changed_paths, tests, findings, and blocker.

Never implement, self-assure, select models, or start another invocation without authority.

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.

Check your setup, free