# Analyze swarm performance bottlenecks

Claude Flow swarms can experience performance issues affecting speed and efficiency. This skill detects bottlenecks across communication, processing, memory, and network layers. It generates detailed performance reports and provides actionable optimization recommendations to improve swarm performance.

## Install

```bash
npx skillstore add claude flow team/ruvnet-performance-analysis
```

## Metadata

- - Slug: ruvnet-performance-analysis
- - Version: 1.0.0
- - Author: Claude Flow Team
- - GitHub username: ruvnet
- - License: MIT
- - Repository: https://github.com/ruvnet/claude-flow/tree/main/.claude/skills/performance-analysis
- - Ref: main
- - Supported tools: Claude, Codex, Claude Code
- - Risk level: safe
- - Risk factors: network, filesystem, scripts, external\_commands
- - Quality score: 68
- - Quality tier: warning
- - Public page: https://skillstore.pages.dev/skills/ruvnet-performance-analysis
- - Manifest: https://skillstore.pages.dev/api/skills/ruvnet-performance-analysis/manifest

## Capabilities

- Detect performance bottlenecks across communication, processing, memory, and network layers
- Generate comprehensive performance reports in JSON, HTML, or Markdown formats
- Analyze agent utilization, task completion times, and coordination overhead
- Provide AI-powered optimization recommendations with estimated improvement percentages
- Apply automatic optimizations with --fix flag for common issues
- Compare current performance with historical data to track improvements

## Use Cases

- Monitor production swarms: Track performance metrics and detect issues in production Claude Flow swarm deployments before they impact users
- Review team performance: Generate performance reports to evaluate team efficiency and identify optimization opportunities across projects
- Optimize swarm efficiency: Fine-tune swarm topology and agent configurations based on bottleneck analysis and improvement suggestions

## Prompt Templates

### Quick health check

```
Run a quick bottleneck detection on the current swarm with default settings
```

### Detailed analysis

```
Generate a comprehensive performance report for the last 24 hours including all metrics and recommendations
```

### Fix issues

```
Detect bottlenecks with a 15 percent threshold and apply automatic fixes to resolve critical issues
```

### CI/CD monitoring

```
Export bottleneck analysis to JSON format and fail if critical issues are detected above 10 percent threshold
```

## Limitations

- Requires Claude Flow to be installed and configured for full functionality
- Does not modify source code directly; only provides recommendations
- Cannot diagnose issues outside of Claude Flow swarm operations
- Performance metrics depend on proper logging and monitoring setup

## Best Practices

- Run bottleneck detection after major changes to identify performance regressions
- Set appropriate thresholds: 10-15% for production, 25-30% for development environments
- Review recommendations before applying --fix to understand the changes being made

## Anti Patterns

- Running with default threshold without considering your specific performance requirements
- Applying --fix without reviewing recommendations first
- Ignoring warning-level bottlenecks that can escalate to critical issues over time

## Security Audit

- - Safe to publish: true
- - Audited at: 2026-01-17T07:45:48.199\+00:00
- - Summary: This skill contains only documentation and usage examples for the Claude Flow performance analysis CLI. All static findings are false positives. The flagged 'weak cryptographic algorithm' patterns are skill tags \(optimization, analysis\) in JSON metadata. The 'external commands' and 'child\_process' patterns are markdown code blocks showing example CLI usage, not executable code. The 'path traversal' patterns are documentation links. No executable code, no data exfiltration, no malicious patterns. Pure documentation skill.

## Stats

- - Views: 273
- - Downloads: 5
- - Favorites: 0
- - Popularity score: 0