Create, evaluate, and refine Claude skills through a structured workflow.
npx skills add https://github.com/tomkis/liftcoach --skill skill-creator-tomkis
Please help me install this Agent Skill. Skill: skill-creator Source: https://github.com/tomkis/liftcoach/tree/main/.agents/skills/skill-creator Command: npx skills add https://github.com/tomkis/liftcoach --skill skill-creator-tomkis
Create, refine, and evaluate Claude skills quickly by providing a structured loop that guides drafting, testing, benchmarking, and optimizing skill descriptions.
Draft an initial skill, write a few test prompts, run evals, review results, and iterate until you converge on a robust skill and a strong trigger description.
High-intent search queries and answers about installing and using this skill.
To build and iterate Claude skills fast, use a structured workflow loop that guides drafting, running evals, benchmarking, and optimizing trigger descriptions until the skill is production-ready.
Skill evaluation and benchmarking for prompt engineering is the process of running controlled experiments to measure trigger accuracy, latency, and reliability across test prompts.
You optimize skill trigger descriptions to reduce false triggers by writing test prompts, running evaluations, reviewing activation accuracy results, and refining the description iteratively.
Yes, you need pyyaml installed to automate skill creation workflows, as it is the required dependency for processing the structured configuration and frontmatter in SKILL.md files.
The best way to structure deterministic tasks in a Claude skill is to place executable scripts in a dedicated scripts/ directory, ensuring repeatable and reliable automation outcomes.
Your Claude skill may fail to trigger accurately if the description lacks optimization, requiring you to run controlled benchmarking experiments to measure activation accuracy and eliminate false triggers.
lucasloutayfDiscover and install modular agent skills using the Skills CLI.
pipixia-labsGuide creation of SKILL.md files and coordinate evals, benchmarks, and iteration.
Speirsy11Automate drafting, evaluating, and iterating Claude skills with SKILL.md.
anymouschinaDraft, refine, and evaluate Claude Skills with SKILL.md and evals.
