agentleFS
QuarktexQuarktex/qcobrain/.claude/skills/agents-sync/SKILL.md

agents-sync

Skill4 starsChanged 4 months agoMITView on GitHub

OPTIONAL (requires a Quarktex API key). Discover and refresh remote agent cards via GET /agents into integrations/quarktex/agents/_agents.md, recording each agent's capability plus a trust tier (read-write | read-only | deny — qcobrain's trust triad, also mirrored into connections.md). New agents default to read-only. Trigger on "sync agents", "what agents can I call", "refresh agent directory", "search the Hive for <capability>". Only runs when quarktex.enabled:true.

  • Reads credentials

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. googlarz/agents-sync★ 5same 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 repository16 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:

    If either is missing, refuse in one line and stop: *"Quarktex is off — enable it and add a key per integrations/quarktex/SETUP.md."*

    line 16
  2. Distinctive:

    Run discovery via the relay (public view.

    line 29
  3. Very specific:

    New agent → default trust: read-only: Never auto-promote. If a row already exists, preserve its existing trust value (it's a deliberate …

    line 40
  4. Very specific:

    deny means never call: Keep denied agents listed (don't delete) so the refusal is explicit.

    line 60

≈ “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

OPTIONAL — requires a Quarktex API key in config.yml; refuse with a one-line setup pointer if disabled.

Inherits references/kernel.md (the provenance contract). Reads references/wiki-spec.md for schema. Wire contract: integrations/quarktex/relay-contract.md. Cache format: integrations/quarktex/agents/_agents.md (SPEC §10).

Gate (check first, every run)

  1. Require quarktex.enabled: true in config.yml and a key in .env (bash integrations/quarktex/relay.sh check).
  2. If either is missing, refuse in one line and stop: "Quarktex is off — enable it and add a key per integrations/quarktex/SETUP.md."

What this does

Refreshes the cache of callable agents so /call-agent has cards to work from. It only touches the discovery cache and the trust column — it does not call any agent and writes nothing to wiki/. (Discovery is metadata, not knowledge: agent cards are not facts about your business, so they never enter canon. A fact only enters the brain when /call-agent actually invokes an agent and /capture sources it.)

Execution

Step 1 — Query the directory

  1. Run discovery via the relay (public view; webhooks are hidden):
    bash integrations/quarktex/relay.sh search "<text>" --capability <cap>   # GET /agents?q=&capability=
    bash integrations/quarktex/relay.sh get qt_yyyyyyyy                       # GET /agents/:id (one card)
    With no arguments, do a broad pull to seed the cache. Each card carries id, name, capability, and example_prompt / example_output (the payload shape /call-agent will build from).

Step 2 — Merge into the cache (preserve human-set trust)

  1. For each card, upsert a row in integrations/quarktex/agents/_agents.md (format in SPEC §10): agent-id | name | capability | trust | example_prompt -> example_output (payload shape).
  2. New agent → default trust: read-only. Never auto-promote. If a row already exists, preserve its existing trust value (it's a deliberate human call) and refresh only the derived fields (name / capability / example shape). Keep the one example row in the file as a format guide.
  3. Drop any agent that has disappeared from the directory — or, if it was used, mark it deny rather than silently deleting (so the decision stays greppable). Don't delete rows with read-write/deny trust without flagging it to the user.

Step 3 — Mirror trust into connections.md

  1. Reflect the reachable-agents summary into connections.md (the registry of systems the brain can reach), so the trust triad is visible in one place alongside every other connection. One line per notable agent or a single "Quarktex Hive (N agents, default read-only)" row — your call, but keep it in sync.

Step 4 — Report

  1. Close with a one-screen summary: N cards synced, M new (defaulted read-only), any trust changes, and which capabilities are now reachable. Point the user at /call-agent to actually invoke one.

Guardrails (hard)

  • Trust is human-owned. This skill may propose a tier for a new agent (always read-only) but must never raise an existing agent's trust on its own. Promotion to read-write is an explicit user decision.
  • deny means never call. Keep denied agents listed (don't delete) so the refusal is explicit.
  • Discovery is not knowledge. Cards go to the cache, never to wiki/. No sources/ row, no notes — those only happen when /call-agent makes a real call.
  • Public view only. You see directory cards; you never see or store another agent's webhook secret.

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.