Skip to main content
Connect Reel25 to AI clients like Claude, Cursor, and VS Code. 26 tools for tracking videos, monitoring accounts, discovering content, and running AI analysis — all through natural language.

Overview

The Reel25 MCP server exposes the same capabilities as the REST API but through the Model Context Protocol (MCP) — an open standard that lets AI assistants use external tools directly. Instead of writing API calls, you describe what you need in plain language and your AI handles the rest. Supported clients: Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and any MCP-compatible client. Prerequisites: Node.js 18+, an active Reel25 plan, and an API key.

Setup

1. Get your API key

Create a key at Settings → API Keys. See the Authentication page for details on key format and security.

2. Configure your AI client

Add to your config file:
  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Windows: %APPDATA%\Claude\claude_desktop_config.json

3. Restart your AI client

After adding the config, restart your application. The 26 Reel25 tools will appear automatically. You can verify by asking: “What Reel25 tools do you have access to?”
Windows users: If you run into issues with npx, try: cmd /c "set REEL25_API_KEY=your_key && npx -y reel25-mcp"

Available tools (26)

Videos — free

Accounts — free

Content Radar — 100 credits

AI Video Analysis — 5 credits

Folders — free

Analytics and Workspace — free

For full credit costs and plan details, see the Credits page.

Async tools

Two tools involve background processing: search_content and analyze_video. Both handle polling internally — when called, the MCP server waits for results and returns the complete response. No extra steps needed from the user or agent.
  • Content Radar (search_content): 30-60 seconds. The system generates search queries, scrapes platforms, filters and ranks results.
  • AI Video Analysis (analyze_video): 15-30 seconds. Downloads video, runs AI models for hook, structure, CTA, and UGC analysis.
Your agent will show a natural “working on it” message during this time.

Environment variables

Example prompts

Discovery and Research
  • “Search for the top UGC skincare product demos on TikTok from the last week”
  • “Find trending fitness content on Instagram Reels about home workouts”
  • “What kind of hooks are working best in the protein supplement niche right now?”
Tracking and Monitoring
  • “Track this TikTok video and add it to my Competitors folder: [url]”
  • “Track these 5 creator accounts and pull their engagement metrics”
  • “Show me all videos I’m tracking that were posted in the last 7 days”
Analysis and Reporting
  • “Analyze this video and tell me what makes the hook effective”
  • “Show me my top 10 performing videos by engagement rate”
  • “Compare growth trends for all accounts in my Influencers folder”
Workspace Management
  • “How many credits do I have left?”
  • “Create a new folder called Q2 Campaign and add these 3 videos to it”
  • “What’s my current plan and how many videos am I tracking?”

Troubleshooting

Restart your AI client completely. On Claude Desktop, quit the app fully (not just close the window) and reopen.
Double-check your API key starts with reel25_sk_ and is set correctly. Regenerate keys at Settings → API Keys if needed.
Ensure Node.js 18+ is installed (node --version to check). On Windows, try the cmd /c wrapper mentioned in the setup section.
Some clients cache tool lists. Remove and re-add the MCP server config, then restart. In Cursor, click the refresh button in MCP settings.
Content Radar and AI Analysis can take 30-60 seconds. If your client has a short timeout, increase it in your MCP config.