
What Does /simplify Do in Claude Code? (Three-Agent Code Review)
The /simplify skill runs three parallel review agents on your changed code — checking reuse, quality, and efficiency — then fixes what it finds automatically.
You've just finished building a feature. The code works, tests pass. But is it clean? The /simplify skill gives you an instant three-agent code review without leaving your terminal — and it fixes what it finds automatically.
How does /simplify work?
When you run /simplify, Claude Code spawns three parallel review agents, each examining your recently changed files from a different angle:
- Code reuse agent — looks for duplicated logic, patterns that should be extracted, and opportunities to use existing utilities
- Code quality agent — checks for common issues like unclear naming, missing edge cases, overly complex conditionals, and style inconsistencies
- Efficiency agent — identifies unnecessary computations, redundant operations, and performance improvements
Each agent reviews independently, then Claude combines their findings and applies fixes.
What makes /simplify different from /review?
The /review command was deprecated in 2026. If you were using it for code reviews, /simplify is its replacement — and it's better:
| Feature | Old /review | /simplify |
|---|---|---|
| Review agents | 1 (single pass) | 3 (parallel) |
| Auto-fix | No (suggestions only) | Yes (fixes automatically) |
| Focus | Correctness | Reuse + Quality + Efficiency |
/simplify doesn't just tell you what's wrong — it fixes what it finds. You review the changes in your diff, not a list of suggestions.
When should you use /simplify?
Run it after completing a feature or fix. The ideal workflow is:
- Build the feature
- Run tests to verify it works
- Run
/simplifyto clean it up - Review the diff
- Ship
This catches:
- Helper functions you duplicated instead of extracting
- Verbose patterns that could be simplified
- Unnecessary intermediate variables
- Logic that already exists elsewhere in the codebase
It's especially useful after AI-generated code, which tends to be functional but verbose.
What does /simplify NOT catch?
/simplify focuses on code cleanliness, not correctness. It won't:
- Verify business logic is right
- Check for security vulnerabilities
- Validate that your feature meets requirements
- Run your test suite
For correctness, run your tests. For security, consider a security review. /simplify is specifically for making working code cleaner.
Can you use /simplify on specific files?
By default, /simplify reviews your recently changed files (based on git status). If you want to target specific files, you can tell Claude:
/simplify Focus only on src/lib/auth.ts and src/lib/session.ts
How much does /simplify cost in tokens?
Running three parallel agents uses more tokens than a single review pass — roughly 3x the cost of a simple review. But it's usually under 50K tokens total, which is well under $1 on API pricing. For the quality improvement, it's worth running on any non-trivial change.