Skip to content

Claude Skills by beyhangl

github.com/beyhangl

1 skillA× 10 installs0 views

EvalcraftAWrite offline, $0 pytest tests for AI agents with evalcraft. Record an agent run to a cassette, generate tests from it, and assert which tools were called, their argument schemas, the output shape, loops, and cost/token/latency budgets. Also check saved cassettes for retired or swapped models. Use when the user imports evalcraft, mentions cassettes or agent regression tests, or asks to test an LLM agent's tool calls, add a cost budget, catch an agent that stopped calling a tool, or detect a r...ai-agentspythonbashVotes: 0GitHub stars: 4

Claude Skills by beyhangl — 1 skill, security-graded | Skills Directory