Model Context Protocol (MCP) Server Directory

Agents need custom models trained on their specific tasks. This FastMCP server exposes Prime Intellect's full RL training pipeline — environment creation, hosted training, evaluation, and 1-click deployment — to any MCP-compatible agent.

Agent commerce needs programmable payment infrastructure. This FastMCP server exposes Stripe Connect's marketplace APIs to AI agents, enabling autonomous vendor onboarding, split payment orchestration, and real-time revenue tracking — all through the Model Context Protocol.

AI agents need observability data to make intelligent operational decisions. This FastMCP server exposes Grafana dashboards, alert rules, and time-series metrics to Claude and Cursor, enabling agents to self-diagnose performance issues and trigger incident response autonomously.

Agents need structured data access to manage projects, track tasks, and coordinate workflows. This FastMCP Python server exposes Airtable's full CRUD API to Claude and Cursor, enabling agents to read, create, update, and search records autonomously through the Model Context Protocol.

Supply chain attacks evolved from compromised npm packages to poisoned MCP servers and tool-description injection. We analyzed 340+ incidents to map the expanded attack surface and build the defense stack agents need in 2026.

Hugging Face released Open-Agent Protocol 1.0 today, an open-source agent communication standard backed by a Linux Foundation working group. OAP 1.0 combines MCP tool discovery, A2A agent-to-agent messaging, and a new Agent Card v2 format with capability negotiation.

Google released Gemini 4.0 Flash today with a 10M token context window, native MCP-compatible tool calling, and $0.15/M input token pricing. The model processes hour-long videos and 1,000-page documents in a single pass, eliminating the chunking and retrieval overhead that has defined RAG architecture since 2023.

Multi-agent systems need reliable event coordination without a centralized orchestrator. This FastMCP Redis Streams server gives agents publish, consume, and acknowledge capabilities with consumer groups that guarantee exactly-once processing across agent fleets of any scale.

AI agents that hardcode API keys create lateral movement risk when compromised. This FastMCP server proxies HashiCorp Vault access so agents receive short-lived, scoped credentials that auto-expire—eliminating stored secrets while maintaining audit trails for every credential access.

AI agents need access to documents, images, and data files stored in object storage—but exposing raw S3 credentials creates a catastrophic blast radius. This FastMCP MinIO server provides scoped, metadata-filtered access with presigned URLs that expire automatically, giving agents safe file access without long-lived credentials.

AI agents managing cloud infrastructure need real-time access to metrics, alarms, and logs. This FastMCP server exposes Amazon CloudWatch to Claude Desktop and Cursor IDE, enabling agents to query metrics, acknowledge alarms, and search logs without leaving their workspace.

Enterprise knowledge trapped in Confluence costs teams 4.7 hours per week in search time. This FastMCP server exposes Confluence pages, CQL search, and space navigation to AI agents, enabling Claude Desktop and Cursor to search, read, and navigate corporate documentation autonomously.