Content
75%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.
A well-structured, highly actionable instruction-only skill: every step is backed by a concrete CLI command, the loop has an explicit termination condition, and validation checkpoints exist. Its main weaknesses are mild redundancy (the context/rules warning stated twice) and the absence of error-recovery guidance for failed commands or writes.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and command-driven with no padding about concepts Claude already knows, but the "context and rules are constraints, do NOT copy them" guidance is repeated nearly verbatim in step 4a and again in the Artifact Creation Guidelines section, which could be consolidated. This matches anchor 4 (efficient with minor instances that could be trimmed) rather than 5, and is well above anchor 3 since there is no unnecessary conceptual explanation. | 4 / 5 |
Actionability | Concrete copy-paste-ready commands are given for every phase (`openspec new change "<name>"`, `openspec status --change "<name>" --json`, `openspec instructions <artifact-id> --change "<name>" --json`) along with the exact JSON fields to parse and the outputPath to write to. Minor gaps (no handling for command failures, no example of an actual artifact write) keep it at anchor 4 rather than 5. | 4 / 5 |
Workflow Clarity | A clearly numbered 5-step sequence with a dependency-ordered creation loop, an explicit stop condition ("Stop when all `applyRequires` artifacts are done"), and validation checkpoints (re-running status after each artifact, "Verify each artifact file exists after writing before proceeding to next") plus a guardrail for pre-existing changes. Validation is present so the batch-operation cap does not apply, but there is no explicit error-recovery loop (what to do when a command or write fails), which keeps it below anchor 5. | 4 / 5 |
Progressive Disclosure | Sections (Input, Steps, Output, Artifact Creation Guidelines, Guardrails) are clearly headed and navigation is easy, and the skill effectively delegates detail to the CLI's `instructions` output at runtime rather than inlining it. At ~90 lines with some repeated guidance that could be consolidated, it sits at anchor 4 rather than 5, which is reserved for a tighter overview with well-signaled split-out references. | 4 / 5 |
Total | 16 / 20 Passed |