agentleFS
tanishq286tanishq286/tq-forge/skills/tq-forge-agents/SKILL.md

tq-forge-agents

Skill8 starsChanged 5 months agoMITView on GitHub

List every forged agent across the sandbox and production with kind, current quality score, hand_off_to wiring, and a one-line capability summary. Read-only. Use when asked for "/tq-forge-agents", "show agents", "list my agents", or "what does this agent do".

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

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

    Suggest follow-ups: only when applicable: sandbox agents >=7 -> /tq-forge-promote.

    line 77

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

/tq-forge-agents — agent registry view

When to use

You want a snapshot of every forged agent — sandbox + promoted — with its kind (researcher / coder / etc.), current quality score, the agents it can hand off to, and a one-line capability summary from AGENT.md. This is the agent-only counterpart of /tq-forge-list, which mixes skills + agents.

Procedure

export TQ_FORGE_HOME="${TQ_FORGE_HOME:-$HOME/.tq-forge}"
source "${CLAUDE_PLUGIN_ROOT:-${TQ_FORGE_HOME:-$HOME/.tq-forge}/install}/scripts/common.sh" && tq_ensure_home
  1. Enumerate agent directories.
    find "$SANDBOX_AGENTS" "$CLAUDE_SKILLS_DIR/agents" -maxdepth 2 -name AGENT.md 2>/dev/null
  2. For each agent, read metadata — yaml + json in one pass:
    python3 - <<'PY'
    import json, re, pathlib, os
    roots = [pathlib.Path(os.environ['SANDBOX_AGENTS']),
             pathlib.Path(os.environ['CLAUDE_SKILLS_DIR'])/'agents']
    for r in roots:
        if not r.exists(): continue
        for d in sorted(p for p in r.iterdir() if p.is_dir()):
            agm, tj = d/'AGENT.md', d/'tools.json'
            if not agm.exists(): continue
            text = agm.read_text(errors='replace')
            kind = (re.search(r'kind:\s*(\S+)', text) or [None,'?'])[1]
            tools = hand = []
            if tj.exists():
                try:
                    j = json.loads(tj.read_text())
                    tools, hand = j.get('tools',[]), j.get('hand_off_to',[])
                except Exception: pass
            loc = 'sandbox' if str(d).startswith(os.environ['SANDBOX_AGENTS']) else 'promoted'
            names = [h if isinstance(h,str) else h.get('name','?') for h in hand]
            print(f"{d.name}\t{kind}\t{loc}\t{len(tools)}\t{','.join(names) or '-'}")
    PY
  3. Pull current scores for each agent slug from skill-log.json:
    python3 -c "import json,pathlib,os;d=json.loads((pathlib.Path(os.environ['TQ_FORGE_HOME'])/'skill-log.json').read_text() or '[]');print(json.dumps({e['slug']:e.get('score') for e in d if e.get('kind')=='agent'}))"
  4. Render the table.
    AGENT               KIND        LOC       SCORE  TOOLS  HAND-OFF
    market-research     researcher  promoted  🟢 8.7    3    business-analyst, reviewer
    inbox-triage        ops-manager sandbox   🟡 6.5    3    coder
    Color: 🟢 >= 8, 🟡 >= 6, 🔴 < 6. Show "—" if score missing.
  5. Print a totals footer.
    <N> agents · <S> sandbox · <P> promoted · avg score <X.X>/10
  6. Suggest follow-ups only when applicable: sandbox agents >=7 -> /tq-forge-promote; promoted agents <7 -> /tq-forge-improve; none found -> /tq-forge-agent <intent>.

Pitfalls

  • An agent with missing or malformed AGENT.md yaml trips the parser. Skip it with a warning rather than aborting the listing.
  • hand_off_to may be a list of strings or objects ([{"name":...,"when":...}]). Show the name field if it's an object.
  • This view reads only forged agents under the two known roots — it won't show unrelated subagents the host may have.

Verification

  • The table has one row per AGENT.md found under the two roots.
  • Total count matches find ... -name AGENT.md | wc -l.

Tags

tq-forge agents registry read-only

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.