Generates structured test plans from requirement, design, or API design documents.
npx skills add https://github.com/HACK-WU/skills --skill test-planner-hack-wu
Please help me install this Agent Skill. Skill: test-planner Source: https://github.com/HACK-WU/skills/tree/main/skills/test-planner Command: npx skills add https://github.com/HACK-WU/skills --skill test-planner-hack-wu
High-intent search queries and answers about installing and using this skill.
Provide the path to your requirement.md and the Skill parses each REQ entry's acceptance criteria, matches keywords to one of seven test strategies, and generates test cases with preconditions, steps, expected results, and pass criteria. Output is written to test-plan.md with a coverage matrix.
In API design mode, the Skill extracts endpoints, request/response schemas, error codes, and auth requirements from api-design.md. Each endpoint gets contract conformance, parameter boundary, error code trigger, and authorization cases, plus idempotency and concurrency cases for write operations.
Seven strategies are matched by keywords: functional, negative, metric, performance, security, data, and experience. Security takes priority when multiple strategies match, and functional requirements always include normal path, exception path, and boundary condition cases.
Yes. It scans the project for existing test directories, config files like pytest.ini or package.json, and framework conventions rather than assuming fixed paths. Each module is annotated with test location, framework, run command, and an executability status such as runnable, environment-dependent, or unknown.
Skip it when you want test code written directly in pytest or jest, when no source document (requirement, design, or API design) exists, or when you only have a quick question about how to write tests. It is designed for full plan generation, not ad-hoc coding.
It runs quality checks confirming every REQ has at least one case or an explicit uncovered reason, P0 requirements are fully covered, and the coverage matrix IDs match actual cases. It also checks that numeric criteria have metric or performance cases and negative criteria have denial cases.
labs21-devDraft a bounded sprint contract from design.md and expose an awaiting_human approval gate.
agenticgogolDrafts test cases from design documents and confirms them with users before implementation.
InNoobWeTrustTrace boundaries, registrations, and dependencies to plan codebase changes.
DozyboyRuns session-based exploratory tests and emits anchored, gate-admissible evidence.
