Assess and validate Test Plans for verifiable behavior and coverage.
npx skills add https://github.com/ylxmf2005/AgentCorp --skill test-plan-reviewer
Please help me install this Agent Skill. Skill: test-plan-reviewer Source: https://github.com/ylxmf2005/AgentCorp/tree/main/agentcorp/test-plan-reviewer Command: npx skills add https://github.com/ylxmf2005/AgentCorp --skill test-plan-reviewer
Judges the quality of Test Plans before implementation to prevent test theater by ensuring the plan contains verifiable, observable checks rather than vague promises. It helps identify gaps where coverage, risk handling, and execution prerequisites are missing and ensures the plan supports measurable outcomes.
Open test/test-plan.md and the associated plan_files to issue a verdict following the four possible outcomes.
High-intent search queries and answers about installing and using this skill.
A test plan review assesses coverage, critical paths, falsifiability, and end-to-end executability against validated requirements to ensure it verifies actual behavior rather than presenting superficial checks.
Test theater occurs when a test plan presents superficial checks instead of verifiable assertions. Prevent it by validating that every objective maps to an observable check, ensuring critical paths and failure modes are explicitly covered.
Validate test coverage by mapping every objective to a verifiable check and tying risk to test coverage. Focus on critical paths, error handling, concurrency, rollback, and data boundaries to ensure the plan supports measurable outcomes.
Yes, you can issue a formal go-no-go decision by analyzing the test plan's executability and context alignment, resulting in a verdict of approve, request_changes, needs_more_evidence, or blocked.
To assess end-to-end executability, you need to provide the test-plan set including test/test-plan.md and plan_files, along with the validated requirements to ensure the plan can run against the declared testing context.
You should request more evidence when the test plan lacks explicit, testable assertions or public contract checks, preventing a clear approve decision and requiring further validation of its observable outcomes.
spandiosGenerate test scenarios from user stories with objectives, conditions, roles, steps, and outcomes.
sunbaoAutomate AIHub QA by running UI and API tests against a remote Docker host.
zxdxjtuGenerate three-layer tests from approved specs into executable test contracts.
thoreinsteinAnalyze Go packages to identify gaps in test coverage without modifying code.
