Content
85%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is highly actionable with executable commands, JSON field contracts, and a well-sequenced workflow including validation feedback loops. Its only weakness is mild redundancy where the context/rules guidance is stated twice.
Suggestions
De-duplicate the 'context and rules are constraints, not file content' guidance: state it once in the Artifact Creation Guidelines and reference it from step 4a instead of repeating it.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is largely lean and action-oriented, but the 'context and rules are constraints for YOU, not content for the file' guidance is restated in both step 4a and the Artifact Creation Guidelines section, so not every token earns its place. | 2 / 3 |
Actionability | Provides fully executable commands (openspec new/status/instructions with --json), names the exact JSON fields to parse, and specifies tools (AskUserQuestion, TodoWrite) and outputPath, making it copy-paste ready. | 3 / 3 |
Workflow Clarity | A clear 5-step sequence with an explicit feedback loop: re-run 'openspec status' after each artifact, confirm every applyRequires artifact is 'done', and verify each file exists before proceeding. | 3 / 3 |
Progressive Disclosure | Self-contained with no bundle files and no nested references; content is organized into well-signaled sections (Input, Steps, Output, Guidelines, Guardrails) appropriate for a single-file workflow skill. | 3 / 3 |
Total | 11 / 12 Passed |