Content
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-crafted, highly actionable analysis skill with a clear multi-step workflow, explicit validation checkpoints, and concrete output formats. Its main weakness is length and some internal repetition—particularly the near-identical pre/post hook logic blocks—which inflates token cost without adding proportional value. The content would benefit from extracting repeated patterns into referenced files.
Suggestions
Extract the duplicated extension hook checking logic (pre and post) into a single referenced file or a reusable template section, referencing it from both steps to reduce ~40 lines of repetition.
Consider moving the detailed detection pass descriptions (Section 4 A-F) into a separate reference file, keeping only a summary list in the main skill body to improve progressive disclosure.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is detailed and mostly necessary given the complexity of the multi-artifact analysis task, but there is significant repetition—the extension hook logic is duplicated nearly verbatim for pre- and post-hooks (Steps 0 and 9), and some sections like 'Operating Principles' restate constraints already covered in 'Operating Constraints'. The semantic model building instructions are appropriately detailed but could be tighter. | 2 / 3 |
Actionability | The skill provides highly concrete, executable guidance: specific shell commands (check-prerequisites.sh with flags), exact file paths, explicit detection categories with examples of what to flag (vague adjectives, placeholders like TODO/TKTK/???), precise output table schemas with column headers, severity heuristics with clear definitions, and specific next-action command suggestions. The report format is copy-paste ready. | 3 / 3 |
Workflow Clarity | The 9-step workflow is clearly sequenced with explicit prerequisites (abort if files missing), progressive loading, multiple detection passes, severity assignment, structured output, and a final remediation offer. Validation is built into the process—Step 1 checks prerequisites and aborts on failure, the read-only constraint prevents destructive operations, and the constitution alignment check serves as a critical validation gate. | 3 / 3 |
Progressive Disclosure | The skill is a monolithic document (~200+ lines) with all content inline. While internally well-structured with clear headers, the duplicated hook-checking logic and the detailed detection pass descriptions could benefit from being split into referenced files. No bundle files are provided to offload complexity, and the skill doesn't reference any supporting documentation for its more complex subsystems (e.g., semantic model building, severity heuristics). | 2 / 3 |
Total | 10 / 12 Passed |