skills/evaluate-team-capability/SKILL.md
Scan linked teammates and score who best fits a user request. Returns ranked agents with relevance scores, reasoning, and a recommendation (delegate, handle-in-main, adapt, or create-new). Call when the topic does not clearly match your active skills or before deciding whether to delegate.
npx skillsauth add bishwashere/cowcode Evaluate team capabilityInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Use this when you are the coordinator (main/CEO) and need to decide whether to handle a request yourself, delegate to a linked specialist, adapt an existing teammate, or offer to create a dedicated agent/mission.
Automatic routing: Before each turn, keyword scoring runs first. If no strong match, an LLM semantic router reads agent titles, aliases, SOUL excerpts, and skills to pick a specialist (when confidence ≥ 0.7). You usually do not need to call this skill manually — it is reflected in inbox routing logs.
agent-send.Do not use for simple greetings or when you already delegated this turn.
The tool returns JSON with:
confidencePct, matchedSkills, reasoningdelegate — strong specialist match; use agent-send nexthandle-in-main — you answer now; offer upgrade if offerUpgrade is trueadapt — partial match on an existing teammate; consider extending their skills or delegating with caveatscreate-new — no good fit; offer a dedicated agent and/or autonomous Mission| action | What to do |
|--------|------------|
| delegate | Call agent-send to the recommended agent (or auto) with the full user message. |
| handle-in-main | Answer the user directly. If offerUpgrade, briefly offer: handle yourself, or create a dedicated specialist / Mission. |
| adapt | Explain the partial fit, delegate with caveats if useful, or offer to broaden a teammate's skills. |
| create-new | Help now if you can, then offer creating a dedicated agent persona + optional Mission for long-term tracking. |
evaluate_team_capability
description: Score all linked teammates (and yourself as coordinator) for a user request. Returns ranked relevance, reasoning, and recommendation (delegate / handle-in-main / adapt / create-new).
parameters:
request: string
tools
Policy-gated command execution for package managers, project generators, build/test scripts, dev servers, and one-off CLIs. Disabled by default; add "exec" to skills.enabled to expose it. Prefer go-read/go-write for stable filesystem primitives.
data-ai
Plain HTTP fetch (GET/POST/etc.) for JSON or text endpoints, including localhost / LAN / Pasture's own dashboard. Use this instead of browse for non-rendered URLs.
testing
Bridge conversation to dashboard Projects and Missions — list configured projects, register new ones with setup details, health-check, propose tasks, create missions after user approval, log progress, and update task status. Use when the user wants to work on, track, or manage a project.
documentation
GitHub integration. Read repos, list/read issues and PRs, create branches, post comments, create PRs. Requires GitHub token in ~/.pasture/secrets.json or GITHUB_TOKEN env var.