Content
71%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.
The body is a compact, well-structured instruction for a single-purpose skill with a clear sequence and correct API-level pointers. Its main gaps are the absence of any executable code example or install command, and mild duplication between the Workflow and Rules sections.
Suggestions
Add a minimal executable test example (import convexTest from convex-test, instantiate with the schema, one t.query assertion) so the guidance is copy-paste ready rather than outline-level.
Replace 'Install convex-test + vitest' with the actual command (e.g. npm install -D convex-test vitest plus vitest.config setup) to close the concrete-command gap.
Merge the Rules bullets into the Workflow steps (or keep only the 'in-memory, not a live deployment' rule) to remove the duplication of auth/error-path and determinism guidance.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and assumes Claude's competence (no library explainers, no padding), but the Rules section restates Workflow steps 3–4 (auth/error coverage and determinism), which is a minor instance of content that could be trimmed — matching anchor 4 rather than the perfectly-lean anchor 5. | 4 / 5 |
Actionability | It names real APIs ("t.run", "t.query/t.mutation", "withIdentity", "t.finishInProgressScheduledFunctions") but includes no executable code example, no install command, and step 2 ("seed via t.run, call t.query/t.mutation, assert") reads as an instruction-level outline rather than copy-paste-ready code — matching the 'concrete guidance but incomplete' anchor 3. | 3 / 5 |
Workflow Clarity | The four-step Workflow is a clear sequence (install → write → cover → run) with the vitest run as an implicit checkpoint, but there is no explicit validate-on-failure/fix/retry loop — matching anchor 4; testing is not a destructive or batch operation so no cap applies. | 4 / 5 |
Progressive Disclosure | The skill is under 50 lines, needs no external reference files (none exist in the bundle), and is well organized into Workflow and Rules sections — per the simple-skill guideline this earns a 5. | 5 / 5 |
Total | 16 / 20 Passed |