Content
82%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.
A highly actionable, well-structured skill body with executable commands and clear feedback loops. The main improvement would be consolidating the scattered numbered workflow steps and trimming a few editorial passages.
Suggestions
Consolidate the open → await → reply → end steps into one numbered sequence so the full workflow reads coherently in a single block.
Trim editorial flourishes ('talks to an empty chair', 'indistinguishable from a broken canvas') to keep the body token-lean.
Consider moving the Mermaid authoring guidance and extended Examples into a reference file to keep SKILL.md a tighter overview.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly lean and assumes Claude's competence, but editorial flourishes like 'the human talks to an empty chair' and 'indistinguishable from a broken canvas' could be trimmed without losing meaning. | 4 / 5 |
Actionability | Copy-paste-ready commands throughout — `ecc-plan-canvas open/await/end`, `--reply`, `typing --state` — with JSON output examples covering the common annotation/verdict/chat cases. | 5 / 5 |
Workflow Clarity | The open → await → reply → end sequence is clear with explicit feedback loops and validation backstops (`pending`, the stop hook), though the numbered steps are scattered across prose sections rather than one coherent sequence. | 4 / 5 |
Progressive Disclosure | Single well-organized SKILL.md with clear sections (When to Use, How It Works, Diagrams, Rules, Examples, Anti-Patterns) and no nested references; over 50 lines so the simple-skill exception does not fully apply, but structure is clean. | 4 / 5 |
Total | 17 / 20 Passed |