MCP Catalogs
Homeobservee screenshot

observee

by observee-ai·41·Score 42

SDK for building AI agents with MCPs, featuring 1000+ integrations, OAuth management and observability tools.

ai-llmdeveloper-toolscommunication
7
Forks
0
Open issues
9 mo ago
Last commit
2d ago
Indexed

Overview

Observee SDK is a comprehensive package that helps developers build AI agents with MCP (Model Context Protocol) integration. It provides three main components: Agents for MCP tool integration with LLM providers (Anthropic, OpenAI, Google), Auth for OAuth authentication across 15+ services, and Logger for structured logging and monitoring. The SDK supports both TypeScript/JavaScript and Python, with clear documentation and examples showing how to integrate with services like Gmail, YouTube, and Linear.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

you:Building AI agents that can access Gmail, Slack, and other services with proper authentication
you:Implementing tool calling for AI applications with pre-built MCP integrations
you:Monitoring and logging AI agent usage and performance in production environments
you:Which LLM providers are supported?
you:How many services does the OAuth integration support?

When to choose this

Choose Observee when you need a comprehensive solution for AI agent development with multiple LLM providers, pre-built MCP integrations, and a managed authentication system.

When NOT to choose this

Avoid Observee if you need self-hosted solutions with no external dependencies, or if you prefer a more minimal MCP server without additional observability features.

Tools this server exposes

3 tools extracted from the README
  • chatWithToolsasync (prompt: string, options: { provider: string, observeeApiKey: string }) => Promise<{ content: string, toolCalls?: any[] }>

    Chat with AI agents using integrated MCP tools for various services

  • callMcpAuthLoginasync ({ authServer: string }) => Promise<{ url: string }>

    Initiate OAuth login for various services

  • Loggernew Logger({ apiKey: string, service?: string }) => Logger

    Create a logger for structured logging and monitoring

Note: Tool names and signatures inferred from usage examples rather than explicit documentation. The README mentions '15+ pre-built tools' but doesn't list them individually. We've included the main SDK functions that appear to be MCP tools.

Comparable tools

mcp-server-examplesnomic-mcpbrowser-tools-mcpfilesystem-mcp

Installation

Installation

**TypeScript/JavaScript:**

# Install everything at once
npm install @observee/sdk

# Or install individual packages
npm install @observee/agents @observee/auth @observee/logger

**Python:**

# Install individual packages
pip install mcp-agents agent-oauth mcp-logger

# Or install all at once
pip install observee

**Claude Desktop Configuration:** Add to claude_desktop_config.json:

{
  "mcpServers": {
    "observee": {
      "command": "python",
      "args": ["-m", "observee_agents"]
    }
  }
}

FAQ

Which LLM providers are supported?
Observee supports Anthropic Claude, OpenAI GPT, and Google Gemini.
How many services does the OAuth integration support?
The Auth component supports OAuth 2.0 flows for 15+ services including Gmail, Slack, Notion, Linear, GitHub, and more.

Compare observee with

GitHub →

Last updated · Auto-generated from public README + GitHub signals.