Content
85%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 strong, executable skill body with a well-sequenced workflow and an explicit validate-and-fix feedback loop before summary. It respects token budget while remaining concrete, with only minor opportunities to externalize reference material.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is largely efficient, leaning on concrete CLI commands and terse bullet rules rather than explaining concepts Claude already knows; a few sentences (e.g. restating the merge intent) could be trimmed but it stays close to the 4 anchor. | 4 / 5 |
Actionability | Provides fully executable, copy-paste-ready commands (`openspec list --json`, `openspec status --change`, `openspec validate`) plus concrete path conventions (`openspec/specs/<capability>/spec.md`) and a worked delta-format reference that covers the common cases. | 5 / 5 |
Workflow Clarity | Steps are explicitly numbered 1-5 with a clear sequence, and validation is built in as a feedback loop (`openspec validate` and 'fix any reported issues before summarizing'), satisfying the destructive/batch validation requirement. | 5 / 5 |
Progressive Disclosure | Well-organized into clear sections (Steps, Delta Spec Format Reference, Key Principle, Output, Guardrails) with one-level structure and no nested references; since there are no bundle files, content is appropriately self-contained, falling just short of a 5 only because a few inline blocks (e.g. the full delta-format reference) could live in a reference file. | 4 / 5 |
Total | 18 / 20 Passed |