{"skill":{"slug":"flow-test","displayName":"flow test","summary":"Designs agent-evaluated flow tests for browser tasks, LLM outputs, and tool workflows. Invoke when exact asserts are brittle and semantic success matters mor...","tags":{"latest":"0.0.1"},"stats":{"comments":0,"downloads":134,"installsAllTime":0,"installsCurrent":0,"stars":1,"versions":1},"createdAt":1774374974364,"updatedAt":1774375323148},"latestVersion":{"version":"0.0.1","createdAt":1774374974364,"changelog":"- Initial release of the \"flow-test\" skill for agent-evaluated testing of browser tasks, LLM outputs, and tool workflows.\n- Provides a structured approach for designing tests where exact asserts are brittle and semantic success is more important than literal equality.\n- Introduces clear guidelines for when to use flow tests versus traditional assertions.\n- Defines an evidence-based evaluation process, including deterministic checks, semantic rubrics, and a bounded verdict system (`pass`, `fail`, `needs_review`).\n- Includes output templates and deliverable requirements for consistent test design.","license":"MIT-0"},"metadata":null,"owner":{"handle":"qipengguo","userId":"s170pr38p1397r6p2bxqgmp6p583hzq7","displayName":"QipengGuo","image":"https://avatars.githubusercontent.com/u/8382210?v=4"},"moderation":null}