skill-pilot
Public UnreviewedLints SKILL.md files for description quality (1,536-char cap, front-loaded triggers, valid frontmatter), and ships a UserPromptSubmit hook that injects a system reminder naming the skills that match e
Skills
The skills this repo ships, by tier, each with a quick health read. Install the user skills with:
adom-wiki skills install kyle/skill-pilotdev-skills/ or publish-skills/, so a contributor (or their AI) has no in-repo guide to build, debug, or publish it. Consider adding them; they stay source-only and never ship in the tarball.Audit Claude Code SKILL.md files and route user prompts to matching skills via a UserPromptSubmit hook so Claude actually invokes the right skill. Trigger words: skill-pilot, lint skills, audit skills, skill description quality, skill front-load, skill triggers, why isn't claude using my skill, skill not triggering, skill routing, prompt routing, route to skill, skill matcher, skill router, enable skill hook, disable skill hook, skill-pilot on, skill-pilot off, skill-pilot status, skill-pilot lint, skill-pilot analyze, analyze skill matches, false positive skills, noisy skills, skill match log, skill router stats, audit my skills, check skill descriptions, are my skills formatted right, claude doesn't use my skills, claude misses skills, skill triggering, 1536 cap, skill cap, system reminder skills, claude code hook skills. Use when the user wants to audit their SKILL.md corpus for description quality (front-loaded triggers, length under the 1,536 cap, valid frontmatter), enable or disable the prompt-router hook that nudges Claude toward matching skills, see what skills matched recent prompts, or troubleshoot why Claude is ignoring a skill it should be using.
No dev skills in this repo.
How to build, debug, and test this app. Source-only (dev-skills/), never shipped in the tarball.
No publish skills in this repo.
The app-to-wiki publish glue. Source-only (publish-skills/), never shipped in the tarball.
Health: the size chip is green when right-sized, yellow when getting long, red when the model likely skims it. A green check is a passed preamble/structure signal; an amber mark is a gentle nudge, not a hard failure.