agentleFS
DeepExtremaDeepExtrema/Tesseract/skills/rounds/SKILL.md

rounds

Skill4 starsChanged 14 days agoMITView on GitHub

Use when Taimoor asks for a vault health check, hygiene sweep, or lint pass — "how's the vault", "run rounds", "anything drifting?", "check the caretakers" — or before trusting the vault after a long gap. Inspects Librarian, graph, organizer, and task health read-only and reports findings worst-first.

The file

Files
SKILL.md

Rounds — caretaker inspection, read-only

Requires the tesseract MCP server. Rounds INSPECT and REPORT; they never fix. Every apply=True is Taimoor's call, made after reading the report — the consent gate is the design, not a formality.

Steps

  1. librarian_status() — last sweep timestamp (older than 48h opens the report with ⚠ stale sweep), per-step failures, health checks (stale embeddings, manifest drift, orphaned entities, cache consistency), pending proposal counts.
  2. graph_stats() for scale, then consolidate_graph(apply=False) — duplicate-entity merge candidates: report the count and the top 3 proposed merges verbatim.
  3. organize_vault(apply=False) — filing proposals: count plus 3 notable examples (note → proposed folder).
  4. list_tasks() — open-task count; flag likely-dead tasks (past-due dates, obvious duplicates) by quoting them, nothing more.
  5. Compose the rounds report in chat, worst first:
    • 🔴 Broken — failed sweeps, health-check failures, tool errors.
    • 🟡 Drifting — stale sweep, duplicate entities, proposal backlog, dead tasks.
    • 🟢 Healthy — one line, counts only. Each finding is one line carrying its evidence. A section with nothing to report says "none" — the eye learns the layout.
  6. For each 🔴/🟡 finding, offer a follow-up; add_task only the ones Taimoor blesses. Applying merges/moves happens only if Taimoor explicitly says so, and then via the same tools with apply=True.
  7. Chat-only output — a health report expires in days; filing it would teach search to retrieve dead state. If Taimoor wants a record, fold it into the day's log_session instead.

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.