Find a tool for each step this project still does by hand, and write the picks into CLAUDE.md under `## Tools` for you to review in the diff.
73%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Critical
Do not install without reviewing
Fix and improve this skill with Tessl
tessl review fix ./sound/skills/stack/SKILL.mdEdit nothing except the project-root CLAUDE.md. Install nothing. Change no manifest.
Read the manifests, scripts, CI, README hand-steps, and the goals in the instruction files. List each step a human still does by hand.
A step whose doing is a decision, or builds the user's understanding, is kept by the user: it gets no tool.
sound:find once per gap, one subagent per gap, all in one message so they run in parallel. Each gets the gap as its need, the project's current tool, and your memory pick.Put one line per gap under ## Tools in the project-root CLAUDE.md, in the working tree: create the file, or merge into the existing ## Tools and keep its lines.
<step>: use <owner/repo>; run <first command> firstA gap where sound:find names no tool gets no line.
The first command is read-only (help, version, skills listing) and runs the installed binary, never npx or curl | sh: later sessions run it unasked, and a tool not yet installed then fails loudly.
Everything else goes in chat only:
End on what is now settled and what is still open.
saadshahd/moo.md
2a844a8
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.