kb-triage
Skill0 starsChanged 31 days agoMITView on GitHub
Use when the student asks to triage, process, file, or empty the inbox ("process my drops", "clean up the inbox"); on demand only, never scheduled. Batch-absorbs inbox/ drops into pages, lists, and reminder ledgers behind safety gates.
In this repository11 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- course-setupSkill
- exam-prepSkill
- kb-captureSkill
- kb-gardenSkill
- kb-ingest-sourceSkill
- kb-secretsSkill
- kb-writeSkill
- setupSkill
- whats-dueSkill
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
- Very specific: line 11
Read the drops as one batch (oldest first, at most TRIAGE_MAX_DROPS_PER_RUN) and cluster related ones before writing anything.
- Very specific: line 14
Write a run report at runs/triage/YYYY-MM-DD-HHMM-triage.md (type: report, from schema/templates/triage-report.md) every run, even when the …
- Distinctive: line 15
Remove a drop only after its content is committed elsewhere.
- Very specific: line 16
Run the checks, re-check the diff cap, commit triage: process inbox batch YYYY-MM-DD.
≈ “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.
Files
kb-triage
AGENTS.md is canonical; this is only a thin task pointer. The full procedure, with every safety gate, is runs/prompts/triage.md. Read it and follow it exactly; it wins over this summary.
- Gates first, before reading any drop:
CLAUDE.local.mdhostname match;runs/PAUSEabsent;git status --shortclean;git pull --rebase(when a remote exists) succeeds; planned changes withinDIFF_CAP_FILES/DIFF_CAP_LINESfromTHRESHOLDS.md. - Read the drops as one batch (oldest first, at most
TRIAGE_MAX_DROPS_PER_RUN) and cluster related ones before writing anything. - A cluster may produce any number of actions: page updates (
kb-writerules), source or research pages (kb-ingest-source), list items, raw artifacts, reminder-ledger bullets, or a question for the student. Do not force one action per drop. - Reminders are recorded as
status: proposed-only·scheduler_ref: noneunless the student asks for a real one in this session (reminders/README.md). - Write a run report at
runs/triage/YYYY-MM-DD-HHMM-triage.md(type: report, fromschema/templates/triage-report.md) every run, even when the inbox is empty. - Remove a drop only after its content is committed elsewhere. Leave blocked drops and say why.
- Run the checks, re-check the diff cap, commit
triage: process inbox batch YYYY-MM-DD. Push only if asked orkb.autopushis on. - Reply briefly: drops processed, pages changed, reminders proposed, questions (at most
NEEDS_OWNER_MAX_QUESTIONS), commit SHA or "no commit".
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
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.

