# MCP Directory Submission Information
## Pre-Submission Checklist
- [x] readOnlyHint added to analyze-video tool
- [x] Repository published to GitHub
- [x] npm package published
- [x] README polished with examples
- [x] SVG logo created (https://github.com/clipsense/-mcp-server/blob/main/logo.svg)
- [x] Privacy policy published (LIVE at https://clipsense.app/privacy)
✅ **ALL REQUIREMENTS MET - READY FOR SUBMISSION**
## Company Information
- **Company/Organization Name**: ClipSense
- **Company/Organization URL**: https://clipsense.app
- **Primary Contact Name**: Geoffrey Buranday
- **Primary Contact Email**: jerlitaburanday@gmail.com
- **Primary Contact Role**: Founder/Developer
## Server Details
- **MCP Server Name**: ClipSense
- **MCP Server URL**: https://github.com/clipsense/-mcp-server
- **npm Package**: @gburanda/clipsense-mcp-server
- **Tagline**: AI-powered mobile debugging through video analysis
- **Description**:
Analyze screen recordings of mobile app bugs with AI. ClipSense processes local video files showing crashes, UI issues, and unexpected behavior in React Native, iOS, and Android apps. Get instant root cause analysis, timeline breakdowns, and code fix suggestions powered by Claude Sonnet 4.5.
## Use Cases + Examples (minimum 3)
### Use Case 1: React Native Crash Analysis
**Prompt**: "Analyze /Users/me/Desktop/app-crash.mp4 and identify why the app crashes"
**Description**: Developer records app crashing when tapping login button. ClipSense identifies null pointer exception in ProfileScreen.tsx and suggests adding null safety checks.
### Use Case 2: UI Glitch Debugging
**Prompt**: "What's causing the layout issue in this video: /Users/me/Downloads/ui-bug.mov"
**Description**: Designer notices profile image overlapping with text. ClipSense identifies missing flex constraints and suggests CSS fixes.
### Use Case 3: Android Performance Issue
**Prompt**: "Analyze /Users/me/Videos/slow-scroll.mp4 and explain the lag"
**Description**: Product manager reports slow scrolling in feed. ClipSense detects excessive re-renders and recommends React.memo optimization.
## Connection Requirements
- Requires API key (obtained via POST request to https://api.clipsense.app/api/v1/keys/request)
- No OAuth required
- Filesystem access required (reads local video files)
## Read/Write Capabilities
- **Read-only**: Analyzes local video files without modifying them
- No write operations
## Third-party Connections
- [x] Connects to ClipSense API (https://api.clipsense.app)
- [x] Uploads videos to Cloudflare R2 (encrypted storage)
- [ ] No other third-party services
## Categories
- Developer Tools
- Debugging
- Mobile Development
- AI/ML Tools
## Authentication Type
- API Key (static)
- API keys obtained via: `curl -X POST "https://api.clipsense.app/api/v1/keys/request" -H "Content-Type: application/json" -d '{"email":"user@example.com"}'`
## Test Account Access
**Testing Account Credentials**:
- Test API Key: [GENERATE TEST KEY]
- Email: test@clipsense.app
**Test Account Setup Instructions**:
1. No account creation needed
2. Use provided test API key in MCP configuration
3. Test with sample videos in `/Users/jerlitaburanday/clipsense-mcp-server/test-videos/`
**Test Data Availability**:
- Sample bug videos available in GitHub repository
- Test API key has 50 analysis quota
## Server Technical Details
**List of Tools**: analyze-video
**Tool Annotations**:
- [x] analyze-video has readOnlyHint: true
- [x] Tool description clearly states it reads local video files
**List of Resources**: None
**List of Prompts**: None
**Data Handling**:
- [x] Videos encrypted during upload (TLS)
- [x] Videos stored encrypted at rest (Cloudflare R2)
- [x] Videos auto-deleted after 24 hours
- [x] No AI training on user data
## Documentation & Support
- **MCP Server Documentation**: https://github.com/clipsense/-mcp-server
- **Privacy Policy**: https://clipsense.app/privacy
- **Support Channel**: support@clipsense.app
## Launch Readiness
- **Server GA Date**: Already in production (GA)
- **Technical Readiness**:
- [x] Tested on Claude Code
- [ ] Tested on Claude Desktop (not compatible due to 31MB file limit)
- [ ] Tested on Claude.ai (filesystem access required, not applicable)
- **Transport Support**: stdio (standard MCP transport)
- **Server Logo**: https://raw.githubusercontent.com/clipsense/-mcp-server/main/logo.svg
## Policy & Compliance
- [x] Complies with MCP security standards
- [x] Complies with MCP safety standards
- [x] Complies with MCP compatibility requirements
- [x] Provides clear error messages and user guidance
## Additional Information
ClipSense is the first MCP server purpose-built for mobile app debugging. It works with all major AI coding assistants (Claude Code, Cursor, Windsurf, Continue.dev, Cline) and supports 9 video formats (MP4, MOV, WebM, AVI, MKV, FLV, MPEG, 3GP, WMV). Videos are processed by Claude Sonnet 4.5 for frame-by-frame analysis.
**Key differentiators**:
- Only tool that analyzes local video files for mobile debugging
- Supports videos up to 500MB (vs Claude Desktop's 31MB limit)
- Works across React Native, iOS (Swift/Objective-C), Android (Kotlin/Java)
- Free tier: 3 analyses/month
- Average analysis time: 2-3 minutes
## Submission Form URL
https://docs.google.com/forms/d/e/1FAIpQLSeafJF2NDI7oYx1r8o0ycivCSVLNq92Mpc1FPxMKSw1CzDkqA/viewform
