Content
100%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
An exemplar technical-reference skill body: lean, dense with MPS-specific knowledge, fully actionable via concrete tool calls and exact directives, with a clear validated workflow and a well-signaled one-level reference structure. No meaningful weaknesses against the rubric.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and specialized — exact cell-model rules, style-item names, and cardinality directives — and assumes Claude's competence without explaining what MPS or projectional editing is, so every token earns its place. | 3 / 3 |
Actionability | It gives concrete, executable guidance: specific tool calls (`mps_mcp_scaffold_editor`, `mps_mcp_check_root_node_problems`, `mps_mcp_insert_root_node_from_json`), exact flags (`type="component"`, `includeComponents`, `readOnly=true`), and named JSON blueprint patterns — fully actionable even though runnable code lives in the reference files. | 3 / 3 |
Workflow Clarity | A numbered workflow (steps 0–5) with explicit validation checkpoints — step 0 checks the editor model already exists via `mps_mcp_get_project_structure`, step 4 inspects with `mps_mcp_print_node`, step 5 validates with `mps_mcp_check_root_node_problems` — matching the clear-sequence-with-validation anchor. | 3 / 3 |
Progressive Disclosure | SKILL.md is a lean overview with a "Reference Index" clearly signaling five one-level-deep reference files (`cell-models.md`, `editor-components.md`, `editor-patterns.md`, `indent-layout-styles.md`, `stylesheet-values.md`), all of which exist in `./references/`, with content appropriately split and easy to navigate. | 3 / 3 |
Total | 12 / 12 Passed |