Content
100%Weight 40%Scale 1-3Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
An exemplary instruction-only skill body: concise, immediately actionable with concrete tool calls and file paths, a clearly sequenced workflow with an explicit conflict checkpoint, and clean section organization appropriate to its size. Nothing is padded or inferred.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and assumes Claude's competence — it never explains what an MCP server, a coding rule, or a filesystem workspace is, and each section earns its tokens; the intro establishes only the non-obvious read-only / single-source-of-truth constraints. | 3 / 3 |
Actionability | Guidance is fully executable: a specific tool call with parameters ('Call mcp__team-rules__file_list with path: "/rules" and depth: 2'), concrete rule-file paths to read per situation, and a quick-reference table mapping situations to files — copy-paste ready, no pseudocode. | 3 / 3 |
Workflow Clarity | The 'Before writing or editing code' section lays out a clear 4-step sequence (list → read applicable rules → cite in plan → apply), and the 'Rules of engagement' section adds an explicit conflict checkpoint ('surface the conflict and ask before deciding') plus a no-applicable-rule guard, providing the feedback loop the rubric rewards. | 3 / 3 |
Progressive Disclosure | At under 50 lines with no external bundle files (references/, scripts/, assets/ are absent) and a single focused purpose, the well-organized sections (intro, before-editing workflow, rules of engagement, quick-reference table) satisfy the simple-skill carve-out for a top progressive-disclosure score. | 3 / 3 |
Total | 12 / 12 Passed |