Applies red-green-refactor to skill writing itself - baseline a subagent without the skill, then iterate until it holds up under pressure.
Visit skill This is mine — claim it
Source: https://github.com/obra/superpowers-skills/tree/main/skills/meta/testing-skills-with-subagents
Listing URL: https://wingmanprotocol.com/skills/testing-skills-with-subagents/
Show your listing is live — the card backlinks here and picks up the verified badge automatically once you claim it.
<iframe src="https://wingmanprotocol.com/embed/card/testing-skills-with-subagents" width="340" height="170" style="border:0;border-radius:10px" loading="lazy" title="Testing Skills With Subagents — live on WingMan Protocol"></iframe>
Or the one-line script:
<script src="https://wingmanprotocol.com/embed/card.js" data-slug="testing-skills-with-subagents" async></script>
Or a README badge:
[](https://wingmanprotocol.com/directory/testing-skills-with-subagents?utm_source=badge)
Sets up hooks, memory scoping, and reflection patterns so a coding agent keeps learning across sessions.
Checks a skills library's links, naming, and cross-references so the collection stays coherent as it grows.
Autonomous spec-driven build system with a built-in trust layer - RARV-C closure loop, quality gates, and a verified-completion evidence gate before it calls wo
Syncs a local skills checkout with new upstream skills from the community repo.
Packages a locally-developed skill and contributes it upstream via a branch and pull request.
Creates, edits, evaluates, and tunes the trigger wording of other Claude skills, including running benchmarks against them.