Run artifact and trigger evals for skills in Docker containers.
npx skills add https://github.com/peshay/portfolixir --skill bmad-eval-runner-peshay
Please help me install this Agent Skill. Skill: bmad-eval-runner Source: https://github.com/peshay/portfolixir/tree/main/.claude/skills/bmad-eval-runner Command: npx skills add https://github.com/peshay/portfolixir --skill bmad-eval-runner-peshay
This Skill provides a clean, isolated environment to run skill evaluations and benchmarks, ensuring results reflect the skill itself without interference from external configurations.
Run evals for the 'bmad-product-brief' skill in a Docker container.
High-intent search queries and answers about installing and using this skill.
Use this Skill to execute skill evaluations in a Docker container, which separates the benchmarking environment from global configurations to ensure results reflect the skill itself without external interference.
Artifact evals execute a skill against a prompt and grade its outputs, while trigger evals measure whether the skill's description triggers Claude on given queries to validate skill performance and correctness.
Yes, Docker is required to run isolated skill benchmarks. The Skill depends on containerized execution to provide a clean environment that separates benchmarking from your local global configurations.
Yes, you can run skill evaluations in a local directory or a Docker container, isolating the benchmarking environment from global configurations to ensure results accurately reflect the skill's own performance.
The best way to validate if a skill description triggers Claude correctly is by running trigger evals, which measure whether the skill's description activates Claude on given queries within an isolated environment.
Isolated skill evaluation requires Docker, Node, Python, jq, and rsync. These dependencies support the containerized execution environment and the scripts needed to run artifact and trigger benchmarks.
diegosanchespereira1Run artifact and trigger evaluations in isolated Docker or local environments.
8b-isDesign and refine Claude Code Skills with consistent SKILL.md structure.
darknetopencv-altGuide creation, testing, and optimization of AI skills with SKILL.md.
lucasolsRun strict TypeScript evals for LLM/agent workflows via CLI.
