Content
77%Reviews 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 with a clear, validated workflow, but it is monolithic and repetitive for its length, with the same guidance restated across several sections.
Suggestions
Consolidate the redundant "graph first, file second" material — the Decision Framework and Anti-Patterns sections overlap heavily with the When-to-Use table; merge or cross-reference to cut length.
Split the detailed per-tool reference (Available MCP Tools) into a one-level-deep reference file and keep SKILL.md as a concise overview to improve progressive disclosure.
Add at least one example MCP tool invocation with arguments so the actionability extends from tool selection to actual call construction.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The "graph first, file second" guidance is repeated across the Core Principle, ASCII box, When-to-Use table, Decision Framework, and Anti-Patterns sections; accurate and relevant but could be tightened substantially by consolidating these overlapping restatements. | 2 / 3 |
Actionability | Concrete tool-to-task mappings (search_graph, get_code_snippet, trace_call_path, detect_changes) with purposes and a copy-paste-ready .mcp.json config provide specific, actionable guidance for tool selection. | 3 / 3 |
Workflow Clarity | The numbered 0–6 workflow (PLAN, LOCATE, UNDERSTAND, BLAST, TRACE, CHANGE, VERIFY) is clearly sequenced with an explicit validation checkpoint (VERIFY via detect_changes) and emphasis ("Never skip step 3"). | 3 / 3 |
Progressive Disclosure | No bundle files exist and the ~150-line body is monolithic; the detailed tool reference, freshness internals, and config sections are inline where a skill this long could split them into one-level-deep reference files. | 2 / 3 |
Total | 10 / 12 Passed |