byelliotjlt· 39 skills
When reviewing, optimizing, or inheriting an LLM agent, systematically identify what's actually load-bearing vs. cruft. Instruments usage, ablates components, and strips the agent to minimum viable form. Answers: "What can we remove without degrading quality?" The goal is the leanest agent that still works.
$npx -y skills add elliotjlt/claude-skill-potions --skill agent-auditInstalls into the current project.
Run `npx skills use "https://github.com/elliotjlt/claude-skill-potions" --skill "elliotjlt/claude-skill-potions/agent-audit"` and follow the generated skill instructions now. Read its complete output, redirecting it to a temporary file first if necessary. Resolve relative paths from the supporting-files directory it provides.
Use the skills in "https://github.com/elliotjlt/claude-skill-potions" that are relevant to the current task. Run `npx skills add "https://github.com/elliotjlt/claude-skill-potions"` and select the relevant skills, then follow their instructions.