building-effective-skills

Human-Frontier-Labs-Inc/human-frontier-labs-marketplace

Meta-skill for building Claude skills that measurably improve output. Gather canonical docs (pith) -> detect where the model drops the ball one-shot AND multi-turn -> distill only the gap -> validate with a blind 4-cell benchmark and a multi-turn drift test. Use when creating/improving a skill, turning docs into a skill, or deciding whether a skill is worth building. Proven: a skill prevented multi-turn drift 13-0 across two stacks.

version

1.0.0

GitHub stars

6

Sponsors

SnugThe clipboard manager you and your agent can use, safelyAgent CaffeineClose your lid, come back to finished workWaxBlock anything with supervision, without erasing your iPhoneOpen StepsSponsoredFree and open-source skills that make AI coding agents easier to understand, verify, and control.