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.
The body is well-structured, executable, and properly split across reference files with a clear validating workflow. Its only weakness is minor over-explanation in a few code comments and bullets.
Suggestions
Trim inline comments that restate general SQL knowledge (e.g. the ANALYZE row-count explanation) to keep only skill-specific guidance like the optimization rationales.
Consider moving the EXPLAIN 'Key things to check' bullets into references/optimization.md to keep the overview leaner.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly efficient with executable examples and tight sections, but a few inline comments and bullets restate concepts Claude already knows (e.g. "PostgreSQL: always use ANALYZE to see actual row counts vs. estimates", "Isolate expensive subquery logic for reuse and readability"). | 4 / 5 |
Actionability | Provides fully executable, copy-paste-ready SQL across the common cases: CTE pattern, window functions, EXPLAIN ANALYZE, and a before/after optimization with a supporting covering index. | 5 / 5 |
Workflow Clarity | The five-step Core Workflow sequences Schema Analysis through Document, with an explicit validation checkpoint in Verify (EXPLAIN ANALYZE, no seq scans, sub-100ms target) and a feedback loop ("if query does not meet sub-100ms target, iterate... before proceeding"). | 5 / 5 |
Progressive Disclosure | A Reference Guide table signals five one-level-deep reference files, all of which exist in ./references/, each with a topic and a "Load When" condition; the body keeps key examples inline and pushes detail to those files. | 5 / 5 |
Total | 19 / 20 Passed |