← ClaudeAtlas

coverage-analyzerlisted

Parses vitest/jest/istanbul coverage JSON, rolls up line/branch/function coverage to the requirement level via RTM, ranks uncovered gaps by risk, and enforces domain-specific thresholds. Gate contract — every P0 requirement has 100% coverage of its mapped lines, overall coverage meets the domain threshold, and no unjustified exclusions. PIPELINE-5 step 5 / PIPELINE-6 step 4.
mytechsonamy/VibeFlow · ★ 0 · AI & Automation · score 75
Install: claude install-skill mytechsonamy/VibeFlow
# Coverage Analyzer ## Phase Contract This skill runs in **TESTING** only. Before any other step, read `vibeflow.config.json`'s `currentPhase`. If it is not TESTING, emit: > coverage-analyzer is for TESTING phase; current is `<phase>`. Run > `/vibeflow:advance` after DEVELOPMENT completes before analyzing > coverage. …and stop. Analyzing coverage mid-DEVELOPMENT against a partial test suite produces a misleading floor; the P0-zero-uncovered invariant is only meaningful after TESTING has run the full suite. An L2 Truth-Execution skill. Line coverage is a lower bound on quality — "did the tests run this line?" — but most teams stop there. This skill adds two things that matter more: 1. **Requirement-level coverage.** It maps lines → files → tests → scenarios → PRD requirements via the RTM, so the report answers "is every P0 requirement's code actually exercised" instead of just "is the file average above 80%". 2. **Risk-weighted gap ranking.** The report ranks uncovered code by the same risk signals `test-priority-engine` uses, so the first thing a reader sees is "this P0 file has 40% churn and 0% branch coverage" not "module X is 78.3%". Coverage is a necessary-but-not-sufficient signal; this skill is where necessity becomes visible. ## When You're Invoked - **PIPELINE-5 step 5** — after `regression-test-runner` + `mutation-test-runner` have produced their reports, before `release-decision-engine` computes its verdict. Coverage is one of the si