Content
86%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 lean, highly actionable single-purpose skill: every step is concrete, sequencing is clear, and safety bounds are sensible. The only improvements are showing the exclude flag syntax and adding an explicit post-index verification step.
Suggestions
Show the exclude flag in a real call (e.g., `ctx_index({ path: ".", exclude: ["node_modules", "dist"] })` and the CLI `--exclude` form) instead of only naming it in the Safety section.
Add an explicit verification step after indexing, e.g., run `ctx_search` for a known snippet and confirm it returns focused results before reporting success.
Replace the placeholder `queries: ["..."]` with one realistic query example so the search command is copy-paste ready.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Every line instructs ("Prefer the `ctx_index` MCP tool when it is available", "Use `path`, not large inline `content`, so file bytes do not enter the conversation") with no concept explanation or padding, matching anchor 5 — lean, efficient, and assuming Claude's competence. | 5 / 5 |
Actionability | Concrete, mostly executable guidance: a complete `ctx_index({ path, source, maxDepth, maxFiles })` call, the CLI fallback `context-mode index . --source project:<name>`, and a `ctx_search` example. It sits at anchor 4 rather than 5 because `exclude`/`--exclude` is named in the Safety section but never shown in syntax, and the search example uses a bare `["..."]` placeholder — minor gaps. | 4 / 5 |
Workflow Clarity | Six clearly sequenced steps with a reporting checkpoint ("Report the indexed source label, file count or section count") and a pre-flight guard ("ask the user before raising `maxFiles` above 500") match anchor 4. It is below anchor 5 because there is no explicit validate-that-search-returns-expected-snippets loop after indexing. | 4 / 5 |
Progressive Disclosure | The body is under 50 lines, needs no external references (none exist in the bundle), and is well organized into Instructions, code examples, and Safety sections, which per the rubric's simple-skill guideline merits a 5. | 5 / 5 |
Total | 18 / 20 Passed |