startupbros-com/skill-tuner
Tells you whether a change to an agent-consumed document actually worked: paired comparison with a confidence interval and a non-inferiority margin, content-hashed provenance with drift detection, and an adversarial defect probe. Reads Anthropic's skill-creator benchmark output and supplies the verdict it stops short of.
Inventory every skill the agent is shown, with its per-turn listing cost, usage count and visibility knob, then decide one skill at a time whether the model should start it on its own. Reads files only; starts on the human's explicit instruction.
Evidence-tagged rules for writing documents agents consume. Use when creating or editing skills, AGENTS.md, or CLAUDE.md.
Find and fix real defects in a skill, AGENTS.md, or CLAUDE.md, then prove the description still routes. Runs the bundled token-spending runner, so it starts only on the human's explicit instruction.