Clarity MCP Bootleg Market
About
# Clarity MCP Bootleg Market Turn browser traffic into tool calls. Smuggle your network requests into an MCP. ## What's This All About? Clarity is where you come when you need to get your hands on some Model Context Protocols (MCPs). Our server connects AI assistants like Claude with web APIs extracted from your HAR…
Explore
- Extracts API endpoints from HAR files automatically
- Generates MCP tools using AI from captured traffic
- Forwards ListTools and CallTool requests to a backend
- Transforms responses to match the MCP protocol
- Requires only an API key for authentication
- Integrates directly with Claude Desktop via stdio
npm install -g @lekt9/clarity-mcp
1. Install the Claude Desktop application
2. Open Claude settings and navigate to the MCP section
3. Add a new MCP server with this command:
npx -y @smithery/cli@latest install @lekt9/clarity-mcp --claude
``
4. Start a conversation with Claude and use the tools provided by your Clarity MCP
-
: API key for authenticating with the Clarity backend
- CLARITY_MCP_ID: Optional MCP ID to specify which MCP to use (useful if you have multiple MCPs)
- NEXTJS_APP_URL`: URL of the Clarity backend (defaults to https://clarity.org/)
npm install
MCPs are the secret sauce that AI agents crave. These Model Context Protocols are powerful tools that agents use to interact with external services and fetch real-time data. Without them, AI is trapped in its own bubble.
Right-click anywhere on the webpage and select "Inspect" or "Inspect Element". Alternatively, use these keyboard shortcuts:
- Chrome/Edge/Firefox (Windows): F12 or Ctrl+Shift+I
- Chrome/Edge/Firefox (Mac): Option+Cmd+I
- Safari (Mac): Enable Develop menu first (Safari > Preferences > Advanced > Show Develop menu), then Option+Cmd+I
Turn browser traffic into tool calls. Smuggle your network requests into an MCP.
What's This All About?
Clarity is where you come when you need to get your hands on some Model Context Protocols (MCPs). Our server connects AI assistants like Claude with web APIs extracted from your HAR files. Think of it as a way to "borrow" functionality from websites and turn it into something AI can use.
How It Works
sequenceDiagram
participant Client as MCP Client (Claude)
participant Server as Clarity Server
participant Backend as Clarity Backend
participant API as Target Web APIs
note over Client,API: HAR File Processing
Backend->>Backend: Process HAR file
Backend->>Backend: Extract API endpoints
Backend->>Backend: Generate tools with AI
Backend->>Backend: Store in database
note over Client,API: Runtime Flow
Client->>Server: Connect via stdio
Client->>Server: ListTools request
Server->>Backend: Forward request
Backend->>Server: Return available tools
Server->>Client: Deliver tools to client
Client->>Server: CallTool request
Server->>Server: Transform tool name
Server->>Backend: Forward to backend
Backend->>API: Execute API call
API->>Backend: Return response
Backend->>Server: Return result
Server->>Client: Deliver to client
Installation
```bash
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
