What problem does it solve?
It helps teams use code coverage as a diagnostic signal for test-suite gaps, while avoiding the common mistake of treating coverage percentages as a quality target (a Goodhart’s Law failure mode).
Core Features & Use Cases
- Coverage-criterion hierarchy: Explains how coverage levels differ in granularity (function, statement/line, branch, decision, condition, MC/DC, path) and what each level can and cannot reveal.
- Floor vs ceiling thinking: Clarifies that uncovered code is reliably “untested,” while covered code is not automatically “verified,” preventing misleading conclusions.
- When granularity matters: Uses safety-critical standards (DO-178C and ISO 26262) as evidence that the right criterion (e.g., MC/DC) depends on the code’s failure modes.
- Practical application boundaries: Provides explicit guidance on what to avoid (e.g., don’t use this skill for choosing test levels or configuring tooling; use the dedicated related skills instead).
Quick Start
Use the test-coverage-strategy skill to explain whether a coverage metric like line coverage or branch coverage is a helpful gap signal for your codebase, and what more appropriate criterion you should consider based on the code’s failure modes.
