ap-mcp
by Hevagog
Overview
Orchestrates LLM-powered tool selection and invocation for natural language interactions with external Python functions.
Installation
just runEnvironment Variables
- OLLAMA_HOST
- OLLAMA_MODEL
- MCP_SERVER_URL
- TOOL_PUBLIC_URL
- TOOL_PORT
- PORT
- FLAVOR
Security Notes
The server's `/register` endpoint allows external tools to register themselves, providing a `base_url` for subsequent tool invocations. If a malicious tool registers with a `base_url` pointing to an internal network resource or a sensitive external service, the MCP server could be tricked into performing Server-Side Request Forgery (SSRF) when invoking that tool. The use of `network_mode: "host"` in Docker Compose further increases the risk by giving containers direct access to the host's network interfaces, potentially exposing local services to compromised containers. While no direct `eval()` or hardcoded secrets were found, the ability for external entities to define proxy targets is a critical vulnerability.
Similar Servers
npcpy
Core library of the NPC Toolkit that supercharges natural language processing pipelines and agent tooling. It's a flexible framework for building state-of-the-art applications and conducting novel research with LLMs. Supports multi-agent systems, fine-tuning, reinforcement learning, genetic algorithms, model ensembling, and NumPy-like operations for AI models (NPCArray). Includes a built-in Flask server for deploying agent teams via REST APIs, and multimodal generation (image, video, audio).
Polymcp
A comprehensive TypeScript framework for building and orchestrating Model Context Protocol (MCP) servers and AI agents, enabling LLMs to intelligently discover, select, and execute external tools.
gofannon
Rapidly prototype AI agents and web UIs, build conversational flows, preview interactions, and deploy agent-driven experiences.
ollama-mcp-bridge
Provides an API layer in front of the Ollama API, seamlessly adding tools from multiple MCP servers so every Ollama request can access all connected tools transparently.