Advisory AOS agent health scan before deployment (manifest, services, immune loop, evidence).
4,050 MCP servers
AI Agent task management system - MCP service
Single MCP server exposing search_tool that routes to 1000+ local MCP tools via semantic search
Personal Model Recommender — find the best model for YOUR tasks
Outside-in validation gate for AI-generated text; a separate model checks LLM output, fail-closed.
Neo MCP server and local execution daemon — runs AI/ML tasks locally, no Python required
Agentic OS MCP Server — 169 tools with on-demand loading (26 core), Streamable HTTP transport, background task engine, browser automation
MCP server for bidirectional AI agent communication via ntfy.sh push notifications. Send messages, receive real-time responses, and chat
Offload cheap work from your AI agent to a local Ollama model, at zero cloud cost.
OmniCoreAgent is an open Python agent harness for building autonomous AI agents that use tools, manage memory and context, coordinate
9-LLM consensus + disagreement scoring + cheapest-route picks to fight hallucinations.
Padlet MCP Server — lets AI agents like Claude Code / Claude Desktop / Cursor directly operate Padlet: create new boards, read walls, post c
Prompt Decomposition Engine as MCP Server - LLM-driven decomposition into structured JSON with Four Directions mapping, ambiguity
An AST-based context router to reduce AI token usage. Provides code skeletons and structural symbols to LLM agents (Cursor, Claude
Context-aware prompt enhancement MCP server
MCP Server for Agent Reputation & Trust Scoring
Human approval for AI agents: a named person approves or rejects, silence is not consent.
MCP server for AI agents to report infrastructure needs they encounter during task execution
Tool selection, Cohere-compatible reranking, and on-chain pricing-truth for agents
MCP server that delegates bulk file reading and doc/boilerplate writing to cheaper OpenAI-compatible models
Governed MCP orchestration, tamper-evident memory, and Claude Code Seatbelt hooks.