Validate a single FR-, AI-, or UJ-item against a running application.
npx skills add https://github.com/wtah/spec-master --skill e2e-item-testing
Please help me install this Agent Skill. Skill: e2e-item-testing Source: https://github.com/wtah/spec-master/tree/main/.claude/skills/e2e-item-testing Command: npx skills add https://github.com/wtah/spec-master --skill e2e-item-testing
This Skill provides a structured approach to validate the behavior of a single FR-xxx, AI-xxx, or user journey (UJ-xxx) against a running application, ensuring acceptance criteria are verified and failures are captured for debugging.
Start the e2e-item-testing workflow by triggering /fix-e2e to spawn the tester against the live solution.
High-intent search queries and answers about installing and using this skill.
End-to-end testing for a single requirement involves spawning an e2e-item-tester agent via the /fix-e2e command to validate FR-xxx, AI-xxx, or UJ-xxx items against your running application. The tester executes items sequentially, verifies acceptance criteria, and persists passing tests for CI/CD traceability.
The best way to validate acceptance criteria is using a sequential e2e-item-testing workflow that generates clear PASS/FAIL reports with context and artifacts. This approach captures errors during live application testing and supports debugging by persisting successful test results for traceability.
Sequential e2e-item-testing captures errors and persists passing tests by running items one by one against a live application. It generates PASS/FAIL results with detailed context and artifacts, ensuring failures are recorded for debugging and successful validations are retained for CI/CD traceability.
Yes, you must have the application running via the provided scripts before executing e2e-item-testing. The e2e-item-tester agent requires the live solution to be active to accurately identify and validate the behavior of single FR-xxx, AI-xxx, or UJ-xxx items.
A limitation of single-item e2e testing is that it evaluates one requirement, AI capability, or user journey at a time rather than running parallel batches. This sequential execution ensures accurate error capture and persistence but may require triggering /fix-e2e repeatedly for multiple items.
Yes, e2e-item-testing supports validating both functional requirements (FR-xxx) and AI capabilities (AI-xxx), alongside user journeys (UJ-xxx). The e2e-item-tester agent verifies the acceptance criteria of these single items against a live application and reports any behavioral failures.
hpsgdWrite Given/When/Then acceptance criteria for user stories.
mvillmowDiagnose CI failures by inspecting logs and reproducing issues locally.
gaiastudio-aiExecute manual software tests and capture run-record evidence.
MadAppGangBatch related operations into single messages for parallel workflow execution.
