Content
77%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 content is highly actionable and well-sequenced with strong validation feedback loops, but it is over-long for a single SKILL.md due to repeated commands and policies that could be consolidated. With no bundle files, progressive disclosure relies entirely on in-document headers rather than external references.
Suggestions
Consolidate the repeated `env status --json`, `update-ai`, and `workflow present` command blocks into a single canonical instance referenced by the sections that need them, to reduce conciseness cost.
Move the long Native MCP routing policy and AI/LangChain node rules into reference files under references/ and link to them from SKILL.md, improving progressive_disclosure and trimming the body.
Deduplicate the sync/push/resolve guidance that appears in both 'Sync Discipline' and 'Conflict Handling' into one section to cut redundant tokens.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is accurate and mostly imperative, but ~522 lines with heavy repetition of the same commands (e.g. `env status --json`, `update-ai`, `workflow present`) and re-stated Native MCP routing policy across sections; not verbose with concepts Claude already knows, so above a 2, but more than minor trimming is possible. | 3 / 5 |
Actionability | Copy-paste-ready `npx` commands with exact flags, a complete TypeScript workflow template, and specific error rules ("stop after two repeated failures") provide fully executable guidance for common cases. | 5 / 5 |
Workflow Clarity | Numbered Bootstrap Order and an 11-step Operating Loop give clear sequences, with explicit validation checkpoints (pull→push --verify→test→inspect executions) and destructive-operation caps that act as feedback loops. | 5 / 5 |
Progressive Disclosure | No bundle files exist and the body is a single monolithic 522-line SKILL.md; it is well-sectioned with headers but content that could be split (AI node rules, Native MCP routing, knowledge commands) is inlined, so it sits above the level-2 'inlined' anchor but lacks one-level-deep external references. | 3 / 5 |
Total | 16 / 20 Passed |