MCP Servers

An MCP server for managing a Letta (formerly MemGPT) server and communicating with its stateful agents. Lets an assistant list and create agents, send them messages, and inspect or edit their persistent memory blocks, which is useful for orchestrating long-lived agents that retain context across sessions.

Provides persistent memory capabilities through a knowledge graph stored in a local JSON file. Enables AI agents to create, read, update, and delete entities and their relationships. Useful for maintaining context across conversations, storing user preferences, and building structured knowledge bases that persist between sessions.

Provides a structured sequential thinking tool through the Model Context Protocol. Enables AI agents to break down complex problems into numbered thought steps, revise previous thoughts, branch into alternative paths, and adjust the total number of steps dynamically. Useful for multi-step reasoning, planning, and analysis tasks that benefit from explicit step-by-step thinking.

Official Anthropic MCP server providing direct access to Claude models through the Model Context Protocol. Enables AI agents to invoke Claude for sub-tasks like summarization, analysis, and code generation within agentic workflows. Supports system prompts, multi-turn conversations, and token counting. Updated for the MCP 2026-07-28 spec: a stateless request/response core that runs on serverless and edge infrastructure, a versioned extensions framework covering MCP Apps (interactive in-conversation UI) and Tasks (long-running work), and authorization hardened to align with production OAuth 2.0 and OIDC identity providers such as Entra and Okta.

Official MCP server for the ElevenLabs audio platform. Gives AI agents text-to-speech, voice cloning, speech-to-text transcription, sound-effect generation, and voice library management. Useful for building voice agents, narration, audiobooks, dubbing, and accessible audio experiences.

MCP server for Replicate, which hosts thousands of open machine learning models behind a single API. Lets AI agents search models, run predictions (image, video, audio, and text generation), poll prediction status, and retrieve outputs. Useful for adding generative media and specialized ML capabilities to agent workflows without managing infrastructure.

MCP server for Langfuse, the open-source LLM observability and prompt management platform. Enables AI agents to fetch and render managed prompts, list prompt versions, and query traces, helping teams manage prompts and inspect LLM application behavior from AI-powered tools.

Arize AI's Phoenix, an open-source AI observability and evaluation platform, exposes a remote MCP endpoint that connects agents directly to a Phoenix instance. Lets assistants query OpenTelemetry traces and spans from LLM applications, inspect datasets and experiments, review evaluation results, and investigate latency or quality regressions.

A lightweight MCP server for a local ComfyUI instance that lets AI agents generate and iteratively refine images, audio, and video. Agents can submit workflows, override prompts and sampler parameters, poll job status, and retrieve outputs, which makes conversational iteration on a local diffusion pipeline practical.

Confident AI's official MCP server, connecting AI tools to the platform behind DeepEval for evaluating, observing, and iterating on LLM quality. Agents can trigger and read evaluation runs, inspect metric scores and failing test cases, manage golden datasets, and review production traces to close the loop between eval results and code changes.

An MCP server for fal.ai's hosted generative model catalog, letting assistants create images, video, music, and audio from natural language. Agents can pick a model, submit a generation request with parameters, poll long-running jobs, and retrieve the resulting media URLs directly in the conversation.

Groq's official MCP server, giving agents access to Groq's high-speed inference stack. Supports fast chat and reasoning completions, vision models for interpreting images, text-to-speech and speech-to-text, batch processing for high-volume workloads, and lookups against Groq's own documentation.

LlamaIndex's official MCP server for LlamaCloud, letting agents query managed retrieval indexes and extract structured data from documents. Each configured index is exposed as its own tool, so an assistant can run semantic search over a specific knowledge base and pull parsed fields out of PDFs and other files.

MindsDB's MCP integration gives agents access to AI workflows, data integrations, and automation capabilities through the MindHub platform.

A self-contained MCP server for managing local Ollama models with no external dependencies. Agents can list installed models, pull or remove them, start and stop the Ollama server, run local completions, and use a model-advisor tool that recommends which locally available model best fits a given task.

Comet's official MCP server for Opik, an open-source LLM evaluation and observability platform. Connects an AI host directly to an Opik workspace so agents can read traces and spans, log evaluation scores, manage prompt versions in the prompt library, and inspect project metrics without leaving the editor.

The official Weights & Biases MCP server for W&B Models and Weave, letting AI assistants query and analyze experiment tracking data in natural language. Agents can look up runs and their metrics, compare sweeps, inspect registered model artifacts, and dig into Weave traces to debug LLM application behavior.