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.
The body is a well-structured, actionable read-only review workflow with concrete commands, explicit scope/validation handling, a severity taxonomy, and a copy-paste report template. It assumes Claude's competence and avoids concept over-explanation; the main room for improvement is tightening a few conditional branches and the constraint bullet list.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is efficient: it provides concrete commands and a reporting template without explaining concepts Claude already knows, with only minor phrasing that could be trimmed (e.g. the long "Do not..." bullet list, though necessary for the read-only constraint). | 4 / 5 |
Actionability | Concrete executable commands appear throughout ("comet status . --json", "git status --short --untracked-files=all", "comet state get <change-name> phase") plus a copy-paste finding format and defined severity levels, with minor gaps where guidance is conditional on comet being installed. | 4 / 5 |
Workflow Clarity | A clear numbered sequence (sections 1-5) with explicit checkpoints (mark missing evidence as "not provided", label incomplete scope, ask the user only on material ambiguity) and a severity checklist; minor validation gaps keep it just below a 5. | 4 / 5 |
Progressive Disclosure | No bundle files exist, and the single self-contained file is well-organized with clear section headers; the one external pointer ("comet-classic/reference/classic-layout.md") is a project file rather than a bundle reference, so structure is good with only minor organization gaps. | 4 / 5 |
Total | 16 / 20 Passed |