Content
65%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 concise, codebase-aware, and gives concrete detect/fix patterns, but the workflow lacks a validation checkpoint before applying fixes and contains a dangling internal reference that breaks navigation.
Suggestions
Add a validation step to the workflow (e.g. 'After applying fixes, run the test suite / typecheck and confirm nothing regressed before reporting done').
Resolve the dangling reference to `.claude/rules/sim-hooks.md` — either add the file to the bundle or inline the relevant State-shape guidance so the pointer is not a dead end.
Tighten anti-pattern #3 by moving the sim-hooks sentinel/useRef detail into a short referenced note, keeping the main rule scannable.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Lean and assumes Claude's competence; the Context section adds genuinely codebase-specific conventions (React Query/Zustand roles) rather than explaining concepts Claude already knows; not 5 because the dense parenthetical in anti-pattern #3 could be trimmed. | 4 / 5 |
Actionability | Provides concrete inline code patterns to detect and fix (`useState(prop)` + `useEffect(() => setState(prop))`, `items.find(i => i.id === selectedId)`); not 5 because guidance is pattern-matching snippets rather than copy-paste executable code covering common cases. | 4 / 5 |
Workflow Clarity | Three sequenced steps are present, but with fix=true the skill applies code changes (a destructive operation) yet includes no validation/verification checkpoint, which per the rubric caps workflow_clarity at 3. | 3 / 5 |
Progressive Disclosure | Sections are well organized and external URL references are clearly signaled one level deep, but the body references `.claude/rules/sim-hooks.md`, a file that does not exist in the bundle, leaving a broken navigation pointer. | 3 / 5 |
Total | 14 / 20 Passed |