| name | Performance Hint |
| description | Notices performance issues when viewing code with nested loops, database queries in loops, inefficient algorithms, or missing caching. Use when user shows code that might have O(n²) complexity, N+1 queries, or obvious performance bottlenecks. |
| allowed-tools | Read |
Performance Hint
Lightweight performance awareness for conversations. Provides quick hints about potential bottlenecks without deep analysis.
When to activate
Activate when you notice in code being discussed:
- Nested loops (O(n²) or worse)
- Database queries inside loops (N+1 problem)
- Sorting or filtering inside loops
- Sequential API calls that could be parallel
- Large arrays without pagination
- Missing memoization for expensive operations
- Event listeners without cleanup
What to do
Briefly mention the issue in 1-2 sentences, then suggest running a command:
Example responses:
"This nested loop creates O(n²) complexity on lines 45-52.
Run `/review src/utils.js` to check for performance optimizations."
"I see a query inside a loop (N+1 problem) on line 28. This will make 100+ database calls.
Run `/deep-review` to analyze performance and get optimization suggestions."
"These API calls are sequential (lines 15-18) but could run in parallel with Promise.all.
Run `/quick-check src/api/` for performance check."
Important rules
- ❌ Do NOT perform full complexity analysis
- ❌ Do NOT benchmark or profile code
- ✅ Only mention obvious issues in current conversation
- ✅ Keep hints brief (1-2 sentences)
- ✅ Always suggest a command for detailed analysis
- ✅ Focus on impactful optimizations, not micro-optimizations
Red flags to watch for
Algorithm complexity:
- Nested loops, especially with array methods like
.find() or .filter()
- Sorting inside loops
Database:
- Queries in
for/forEach loops
- Missing
include or eager loading
- No pagination on
.findAll()
Memory:
- Event listeners without removal
- Timers/intervals not cleared
- Large objects in closures
Network:
- Multiple
await calls that could use Promise.all
- No request debouncing
- Fetching entire datasets
Commands to suggest
/quick-check - for fast performance scan/review [file] - for specific file analysis/deep-review - for comprehensive performance review with optimization strategies
This is a hint tool, not a replacement for proper performance profiling.
