coverage-verify

Verify acceptance criteria coverage and generate a traceability report from repository tests.

Updated May 8, 2026
One-click install
npx skills add https://github.com/juanca202/sdd-devkit --skill coverage-verify-juanca202
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: coverage-verify
Source: https://github.com/juanca202/sdd-devkit/tree/main/skills/coverage-verify
Command: npx skills add https://github.com/juanca202/sdd-devkit --skill coverage-verify-juanca202

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) and assets (resource) components.

What problem does it solve? It answers whether implemented code actually covers every acceptance criterion of a user story, work item, or feature, producing an auditable traceability matrix instead of relying on manual checks. ## Core Features & Use Cases - Criteria-to-test traceability: Maps each acceptance criterion (verbatim identifiers like AC-001, 1.1, R-3) to test cases (TC-XXX) and automated test artifacts, classifying each as COVERED, PARTIAL, or UNCOVERED. - Delegated test execution: Reuses fresh quality-check run results or triggers a tests-only run, never executing the suite itself, and reports PASS/FAIL/NOT_RUN per matrix row. - Idempotent reporting: Generates a coverage.md report with a verdict (APPROVED, APPROVED_WITH_NOTES, REJECTED) and fingerprint markers so unchanged work is not revalidated. - Use Case: Before merging a branch for US-042, ask the agent to validate its coverage; it reads the story's criteria and test-cases folder, cross-references repository tests, and saves a coverage.md verdict next to the artifact. ## Quick Start Ask the agent to verify that the code covers the acceptance criteria of US-042 and generate the coverage report.

Frequently Asked Questions about coverage-verify

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I verify acceptance criteria coverage for a user story?▼

Point the skill at the story identifier (e.g., US-042). It reads the README's acceptance criteria, maps each to test cases and repository test artifacts, obtains results from quality-check, and writes a coverage.md report with a verdict.

How to generate a requirements traceability matrix from tests?▼

The skill builds a traceability matrix with one row per criterion, test case, and declared test type, showing evidence path, execution source, and result. It saves the matrix inside coverage.md next to the artifact being validated.

Does coverage-verify run the test suite itself?▼

No, it never executes tests directly. It reuses a fresh quality-check-run.json cache or delegates to the quality-check skill in tests-only mode, and marks results NOT_RUN when execution is not possible.

What happens if a work item has no acceptance criteria?▼

The skill blocks and does not generate a report. Without criteria carrying identifiers there is nothing to trace, so it reports the gap and suggests defining acceptance criteria first.

Can it revalidate archived or already integrated work?▼

Yes, it resolves artifacts under docs/specs/archived/ when they are no longer in the active folder and writes the coverage.md report alongside the archived artifact. Revalidating integrated work is a supported case.

Why does a criterion show PARTIAL when a unit test exists?▼

A criterion is PARTIAL when part of its declared test intent is unmaterialized, such as a test case declaring Unit and E2E where only the unit test exists. It also applies to draft test cases or suite failures that cannot be isolated to a specific test.