Content
92%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
An exemplary instruction-only skill body: a tightly scoped RED-GREEN workflow with explicit validation gates, lean token-efficient prose, and a single well-placed one-level reference. The only improvement area is including literal executable commands for the build/test steps rather than naming them.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and directive throughout — every sentence instructs, nothing explains concepts Claude already knows, and it delegates authority to AGENTS.md instead of restating it. Every token earns its place. | 5 / 5 |
Actionability | Guidance is concrete and specific: exact paths ("docs/openapi/paths.txt", "docs/openapi/latest.json"), exit code `2`, worktree-specific binary paths, and named validation gates. It stops short of a 5 because no literal copy-paste commands are given for key steps (e.g., how to run the focused test or build the binary), leaving minor gaps in executability. | 4 / 5 |
Workflow Clarity | The sequence (design note → RED → API validation → implement → GREEN/verify → handoff) is clearly ordered with explicit validation checkpoints and feedback loops: "Run the focused test and record the expected failure before implementation" and "Rerun the focused failing test after each small fix". | 5 / 5 |
Progressive Disclosure | The ~44-line body is well-sectioned and holds only overview-level guidance, with a single clearly signaled, one-level-deep reference (references/test-matrix.md, verified to exist and contain the promised test cases). Navigation is easy and nothing is nested or buried. | 5 / 5 |
Total | 19 / 20 Passed |