Infino CLI & MCP Server

MCP server that makes your Infino data searchable by AI agents. Automatically captures context: logs, metrics, code changes, and chat history.

🚀 Quick Start (2 minutes)

1. Install

npm install -g infino

2. Set Credentials

export INFINO_ACCESS_KEY="your_access_key"
export INFINO_SECRET_KEY="your_secret_key"

# Save credentials (so you don't need to export them again)
infino creds

3. Test Connection

infino whoami

4. Use with AI Assistants

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "infino": {
      "command": "npx",
      "args": ["-y", "infino"]
    }
  }
}

Restart Claude Desktop → You can now query your Infino data!

Cursor

Add to ~/.cursor/mcp.json:

{
  "mcpServers": {
    "infino": {
      "command": "npx",
      "args": ["-y", "infino"]
    }
  }
}

Restart Cursor → Infino tools are now available!

📖 What Can You Do?

Once connected, your AI assistant can:

• Query your data: "Show me errors from the last 24 hours"
• Analyze logs: "What's causing these timeout errors?"
• Manage datasets: "List all my datasets"
• Upload data: "Upload this chat history for analysis"
• Create visualizations: "Show CPU usage trends"

💻 Installation Options

Recommended: Global Install

npm install -g infino
infino --version

Alternative: Use Without Installing

npx infino whoami
npx infino capture start

Project Dependency

npm install --save-dev infino
npx infino --version

Troubleshooting: If infino command not found after global install:

• Use npx infino instead, or
• Add npm global bin to PATH: export PATH="$(npm config get prefix)/bin:$PATH"

🔐 Authentication

Priority Order

1. Saved credentials: ~/.infino/credentials.json (created by infino creds)
2. Environment variables: INFINO_ACCESS_KEY and INFINO_SECRET_KEY

Managing Credentials

# Save credentials from environment variables
export INFINO_ACCESS_KEY="your_key"
export INFINO_SECRET_KEY="your_secret"
infino creds

# Check current auth status
infino whoami

# Optional: Set custom endpoint
export INFINO_ENDPOINT="https://api.infino.ws:443"

📦 Context Capture

Automatically capture and analyze development context:

Basic Usage

# Start capturing (runs in background)
infino capture start

# Check what's being captured
infino capture status

# Stop capturing
infino capture stop

What Gets Captured?

• Docker logs from running containers
• System metrics (CPU, memory, network)
• Git changes and workspace activity
• Terminal commands (with setup)
• AI chat history (Cursor, Claude, Copilot)

Advanced Options

# See all available collectors
infino capture list

# Capture only specific data
infino capture start --only workspace,docker

# Exclude sensitive data
infino capture start --exclude terminal,ai-chat

# Debug mode (see what's happening)
infino capture start -f

Add to your ~/.zshrc or ~/.bashrc:

export PROMPT_COMMAND='echo "$(date +%s): $(history 1)" >> ~/.infino/context/terminal/shell_$(date +%Y%m%d).log'

🛠️ CLI Commands

| Command | Description | |---------|-------------| | infino | Start MCP server (for AI assistants) | | infino whoami | Check authentication status | | infino creds | Save credentials to file | | infino capture start | Start context capture | | infino capture stop | Stop context capture | | infino capture status | Check capture status | | infino capture list | List available collectors | | infino version | Show version | | infino help | Show help |

🤖 MCP Tools Available

Your AI assistant can use these 18 tools:

Query & Search

• infino_query_natural_language - Natural language search
• infino_chat_history - Search chat transcripts
• infino_query_sql - Execute SQL queries

Dataset Management

• infino_list_datasets - List ALL datasets (local Infino + remote connections)
• infino_get_dataset_metadata - Get dataset info
• infino_get_dataset_schema - Get dataset structure
• infino_create_dataset - Create new dataset

Data Upload

• infino_upload_json - Upload JSON data
• infino_upload_metrics - Upload metrics
• infino_upload_cursor_chat - Upload chat history
• infino_upsert_sql - Upsert data via SQL

Thread Management

• infino_list_threads - List AI conversation threads
• infino_create_thread - Create new thread
• infino_get_thread - Get thread details
• infino_update_thread - Update thread
• infino_delete_thread - Delete thread
• infino_delete_thread_messages - Clear thread messages
• infino_send_message - Send message to thread

🔍 Advanced Features

• infino://config - Current configuration
• infino://recent-queries - Query history
• infino://datasets - Dataset metadata

• debug-errors - Guided error debugging
• code-review - Code change review
• performance-analysis - Performance insights
• chat-summary - Conversation summaries

Enable detailed logging:

{
  "mcpServers": {
    "infino": {
      "command": "npx",
      "args": ["-y", "infino"],
      "env": {
        "INFINO_MCP_DEBUG": "1"
      }
    }
  }
}

📚 Learn More

• Repository: github.com/infinohq/infino
• Website: infino.ai
• MCP Docs: modelcontextprotocol.io

🆘 Support

• Issues: GitHub Issues
• Status: status.infino.ws

📄 License

MIT