Content
78%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 tight, well-structured spec with concrete data contracts and a clear completion criterion, appropriate for a single-purpose extraction atom. The main gap is the lack of a worked example and an explicit validation/feedback loop.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and assumes Claude's competence (no basic-concept explanations), with only minor trimming candidates like "the input shape is intentionally generic so the same atom serves both code-migration and figma-migration". | 4 / 5 |
Actionability | Provides a concrete output file tree, a jsonc token-entry schema with field semantics (sources[]/usage[] audit trail), concrete input sources, and an implementation path; minor gaps are the absence of a fully populated tokens.json example or runnable commands. | 4 / 5 |
Workflow Clarity | The single-purpose action is unambiguous and the Convergence section gives an explicit completion criterion plus empty-bag handling, though there is no validate→fix→retry feedback loop (not strictly required for read-only extraction). | 4 / 5 |
Progressive Disclosure | Under 50 lines with no bundle files needed, organized into clear, well-signaled sections (Inputs, Output, Convergence, Anti-patterns, Status), meeting the simple-skill exception for a top score. | 5 / 5 |
Total | 17 / 20 Passed |