# Optimize

> Review code for efficiency and performance. Triggered by "/optimize" when user wants to identify bottlenecks or improve performance.

- Skill: `x-cmd/optimize` (Agent Skill)
- Install (CLI): `npx skillmds@latest add x-cmd/optimize`
- Raw SKILL.md: https://api.skillmd.com/api/skills/x-cmd/optimize/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: x-cmd (https://skillmd.com/u/x-cmd)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/x-cmd/optimize

---


# Performance Optimizer

## Trigger

Use when:
- User says `/optimize`
- User reports slowness or high resource usage
- User asks for performance improvements

Do NOT use when:
- User asks for readability improvements (use `/simplify`)
- User asks for security audit (use `/security-review`)

## Steps

### Step 1: Profile (if applicable)

If user provides runtime data or if profiling is straightforward:
```bash
# Examples by platform/language
/usr/bin/time -v ./script.sh        # Linux
python -m cProfile -s cumtime script.py
node --prof script.js
```

If no runtime data, skip to Step 2 with static analysis.

### Step 2: Identify Bottlenecks

Analyze for:
- Algorithmic complexity (O(n²) → O(n log n))
- Synchronous I/O in loops
- N+1 query patterns
- Missing database indexes
- Unbounded memory growth
- Repeated expensive computations without caching

### Step 3: Generate Report

Use the output format below.

## Output Format

```markdown
## Performance Review

### Bottlenecks Found

#### 🔴 [Severity] [Category]
**File**: `path:line`
**Issue**: [Description]
**Impact**: [Quantify if possible]
**Fix**: [Recommendation]

#### 🟠 ...

### Recommendations (Priority Order)
1. [Highest impact fix]
2. [Next fix]
```

## Constraints

- Do NOT optimize without profiling data unless obvious algorithmic issue
- Focus on the 20% causing 80% of slowdown (Pareto)
- Verify each suggestion is measurable
- Do NOT auto-apply changes