Content
92%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 tight, well-organized body for a simple single-action skill: it documents the tool call, argument mapping, defaults, scope semantics, and edge guidance (never invent a run_id) with zero padding. The only real gap is the absence of a concrete example tool call showing the parameter structure.
Suggestions
Add one example `search_memories` invocation showing the parameter names and types (e.g., how the user's question maps to the query parameter and how top_k/category/scope/run_id are passed), closing the actionability gap.
State explicitly which parameter carries the user's question (e.g., 'pass the question as `query`') so the flag-to-argument mapping is fully unambiguous.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The ~20-line body is lean and assumes Claude's competence: every sentence carries non-obvious detail (defaults for top_k/category/scope, the meaning of repo/dir/mine scopes, "It filters the memories returned; it does not identify the session making the search request", "Use a known session ID, never invent one"). Nothing explains concepts Claude already knows, matching 'every token earns its place'. | 5 / 5 |
Actionability | Guidance is concrete and mostly executable: the exact tool name (`search_memories`), how to map CLI flags (`--top-k`, `--category`, `--scope`, `--run-id`) to tool arguments, defaults when omitted, and the final instruction "Return the tool's result directly". It falls short of a 5 only because no example call shows the actual parameter names/signature (e.g., how the question maps to a `query` parameter), leaving a minor gap. | 4 / 5 |
Workflow Clarity | This is a simple single-task skill and the single action is unambiguous: call `search_memories` with the user's question, pass flags as arguments, return the result directly — the rubric's simple-skill exception applies. There are no destructive or batch operations, so no validation checkpoint is required for a 5. | 5 / 5 |
Progressive Disclosure | At under 50 lines with no need for external references (no references/, scripts/, or assets/ bundles exist), the single section header and short well-organized paragraphs fully satisfy the rubric's simple-skill criterion. There are no nested or buried references to penalize. | 5 / 5 |
Total | 19 / 20 Passed |