When to Apply This Skill
- A developer asks "what am I missing?" after initial test generation
- A coverage report (coverage.py, Istanbul, JaCoCo) is available and needs interpretation
- Pre-sprint planning requires a gap estimate and effort forecast for a module
- A team wants to know exactly what it takes to reach a coverage target (e.g. 80%)
Atomic Skills to Load First
Read this file before executing any step:
../unit-test-analyzing-code-coverage/SKILL.md— All four coverage types (line / branch / function / statement), gap severity classification, tool integration guides (coverage.py, Istanbul, JaCoCo, coverlet), anti-pattern detection, and all three output report formats
Execution Steps
Read references/index.md before executing any step.
Step 1 — Receive Input
Accepted inputs (one or more):
- Coverage report: JSON (coverage.py / Istanbul), XML (JaCoCo / Cobertura), HTML
- Source code file (for structural gap analysis when no report is provided)
- Existing test file (to cross-reference against source for uncalled paths)
- Coverage target, e.g. "80% line, 75% branch" (defaults to these if not specified)
Step 2 — Analyze
Follow the full analysis workflow from ../unit-test-analyzing-code-coverage/SKILL.md:
- Parse coverage data or analyze source structure statically
- Identify: uncovered lines, untaken branches, uncalled functions, unhandled exceptions
- Classify every gap: CRITICAL | HIGH | MEDIUM | LOW
- Flag coverage anti-patterns (over-permissive mocks, integration tests masking gaps)
Step 3 — Produce Outputs
Generate all three deliverables defined in the unit-test-analyzing-code-coverage skill.
Output Deliverables
coverage_gap_report.md <- prioritized gap table (severity, location, description)
recommended_tests.md <- specific named tests to write, with descriptions
coverage_delta_estimate.md <- projected coverage % once recommended tests are added