The description states a methodology ('red/green/refactor') but does not list concrete actions the skill performs. It reads more like a rule or constraint than a description of capabilities.

The 'what' is vaguely implied (enforce TDD) but not clearly articulated as specific actions. There is no 'when' clause or explicit trigger guidance, which per the rubric caps completeness at 2, and the weak 'what' brings it to 1.

Contains 'TDD', 'red/green/refactor', and 'code change' which are relevant keywords a user might use, but misses common variations like 'test-driven development', 'unit tests', 'write tests first', or 'test before code'.

TDD is a fairly specific methodology, giving it some distinctiveness, but the phrase 'every code change' is broad enough that it could conflict with other coding workflow or testing skills.

Every section earns its place. No unnecessary explanations of what TDD is or why it matters — it jumps straight into the rules and loop. The failure modes table is dense and useful. No padding.

Each step has concrete, specific instructions: run the test, confirm failure, write minimum code, commit. The failure modes table provides exact corrections. The checkpoint format gives a copy-paste template for tracking. This is an instruction-only skill that doesn't need code examples.

The 4-step loop is clearly sequenced with explicit validation at each phase (confirm failure, confirm green, re-run after refactor). The 'Resuming Mid-Session' section handles error recovery. The checkpoint format creates a feedback loop for tracking compliance.

For a skill under 50 lines with no need for external references, the content is well-organized with clear sections (Rule, Loop, Failure Modes, Resuming, Checkpoint Format). Each section serves a distinct purpose and is easy to navigate.