# AI Agent Tools With MCP Support

> Source: https://theagentsindex.com/collections/ai-agent-tools-with-mcp-support (curated, quality-gated, re-verified)

MCP (the Model Context Protocol) is becoming the default way an AI agent reaches your tools and data, instead of a bespoke integration per pair. See our full explainer for what it is and why it matters when you're picking an agent to build on. This collection answers the practical question that piece leaves open: which tools in our own index actually document it today.

61 of the 126 tools here do, spanning every category in the index. Some support it as a client, calling out to any MCP server; some as a server, letting other agents call into them; several as both, including Glean, which ships MCP in both directions today.

The split isn't even. Agent frameworks and coding agents are the densest slices (13 of 16 and 20 of 27), unsurprising since MCP began as a developer-tooling protocol: Google ADK, Claude Agent SDK and Agno all document it too. Sales and marketing agents are where MCP has landed least: only Clay (a full MCP server) and 11x (one undetailed pricing-page bullet, the thinnest evidence in this set) document anything at all.

Not everything in the index is here, and the absences are not all the same kind. Some carry a documented reason on their own listing: Aider, Lindy, GC AI, Artisan, Qualified, Rox, Decagon, Bland AI, Juggler, Zot, Unify, Moveworks and Vecbase. Either they have confirmed they don't support it yet, or no MCP claim appears anywhere in their own docs, site or changelog as of this check. Aider has an open, unresolved GitHub request since December 2024; Lindy's own blog says it isn't compatible yet; Moveworks' own blog still frames MCP as "in development," not shipped.

Undermind no longer fits that group: its own pricing page names MCP support on every tier ("Free at $0 with MCP"), and it now sits among the entries below rather than this absence list.

AgentsKit is an absence of a third kind: its registry exposes an /api/mcp discovery endpoint, which indexes its own catalogue rather than acting as an MCP client or server, so it does not clear the bar this page sets. The others have not been read for MCP support since they were added, so their absence records an unreviewed listing rather than a confirmed gap: Amplemarket, Auto-Respond, Browserbase, Deepgram Voice Agent API, Genspark, Julius AI, Junie, LangGraph, Muse Code, OpenClaw, SkipCalls, Synthflow AI, Warmly and Zendesk AI Agents. A listing published since that check is in neither list yet.

| Name | What it is | Why it's here | Pricing | API |
| --- | --- | --- | --- | --- |
| [Cursor](https://theagentsindex.com/cursor.md) | Coding agent in a desktop editor, terminal CLI, cloud VMs and CI, priced per seat plus token usage. | MCP servers and tools are a core integration surface built into the editor. | — | Yes |
| [Claude Code](https://theagentsindex.com/claude-code.md) | Agentic coding tool from Anthropic that edits your codebase, runs commands, and opens pull requests. | MCP servers and external tools are a core, first-class way to extend what it can reach. | — | Yes |
| [Cline](https://theagentsindex.com/cline.md) | Open-source coding agent for VS Code, JetBrains, the terminal and CI, running on whichever model you bring. | First-class support plus its own in-app MCP marketplace for one-click server installs. | — | Yes |
| [Devin Desktop](https://theagentsindex.com/windsurf.md) | AI IDE with a Kanban command center for running local and cloud coding agents; the renamed Windsurf. | MCP servers are a core integration, carried over unchanged into the Devin Desktop rebrand. | — | Yes |
| [OpenCode](https://theagentsindex.com/opencode.md) | Open source AI coding agent for the terminal, desktop, browser and IDE, running any model you connect. | A dedicated MCP-server integration layer sits alongside its own custom agent-skills system. | — | Yes |
| [Bolt](https://theagentsindex.com/bolt-new.md) | Browser-based AI agent that builds, hosts and publishes full-stack JavaScript web and mobile apps from prompts. | Connects out to MCP servers (Notion, Linear, GitHub, custom) as a client: no server of its own. | — | No |
| [Codex CLI](https://theagentsindex.com/codex-cli.md) | OpenAI's terminal coding agent for inspecting, editing and running repository code from the shell. | MCP servers and tools are documented alongside its sandboxed execution model. | — | Yes |
| [Devin](https://theagentsindex.com/devin.md) | Cloud AI software engineer that takes a ticket, works in its own VM, and opens a pull request. | An MCP Marketplace ships as a client, connecting Devin to external MCP servers. | — | Yes |
| [GitHub Copilot](https://theagentsindex.com/github-copilot.md) | GitHub's AI coding assistant and cloud agent, from editor completions to pull-request review. | Both Copilot Chat and the autonomous coding/cloud agent connect out to external MCP servers. | — | Yes |
| [OpenHands](https://theagentsindex.com/openhands.md) | Open-source software engineering agent with a browser client, CLI and Python SDK; local, cloud or self-hosted. | An MCP client over SSE, Streamable HTTP or Stdio transports, configured directly. | — | Yes |
| [Replit Agent](https://theagentsindex.com/replit-agent.md) | AI agent that writes, tests and publishes full apps from a chat prompt in Replit's browser workspace. | Documented as an MCP client in Replit's own developer docs. | — | Yes |
| [Factory](https://theagentsindex.com/factory-ai.md) | Droid coding agents that run in the desktop app, CLI, IDEs, CI and Factory-managed cloud computers. | Droid is an MCP client with a built-in registry of 40+ pre-configured servers, browsable via `/mcp`. | — | Yes |
| [CodeRabbit](https://theagentsindex.com/coderabbit.md) | AI code review across pull requests, IDE and CLI, with agentic fixes, pre-merge checks and security review. | Connects out to MCP servers as a client only: plan tier caps how many, from 5 on Pro up to 20 on Enterprise. | — | Yes |
| [Antigravity CLI](https://theagentsindex.com/antigravity-cli.md) | Google's terminal agent: a keyboard-driven TUI that edits code, spawns subagents and gates every tool call. | Ships its own local MCP server exposing tools and prompts, plus first-party servers for Google Workspace and Data Cloud. | — | Yes |
| [Warp](https://theagentsindex.com/warp.md) | Agentic development environment: a terminal, a standalone agent CLI, and unattended cloud coding agents. | Its Oz agent CLI takes MCP server config directly (by UUID, inline JSON or file path) to reach tools like GitHub, Linear and Sentry. | — | Yes |
| [LangChain](https://theagentsindex.com/langchain.md) | Open source agent frameworks plus LangSmith, the platform for tracing, evaluating and deploying agents. | The official `langchain-ai/langchain-mcp-adapters` package makes LangChain agents a fully-supported MCP client. | — | Yes |
| [CrewAI](https://theagentsindex.com/crewai.md) | Open-source multi-agent framework with a managed platform for deploying and governing agent workflows. | An official `MCPServerAdapter` connects crews to external MCP servers as a client. | — | Yes |
| [LlamaIndex](https://theagentsindex.com/llamaindex.md) | Open-source agent framework plus a credit-metered document parsing, extraction and indexing platform. | Both directions: consumes external MCP servers and can publish its own workflows as one. | — | Yes |
| [OpenAI Agents SDK (Python)](https://theagentsindex.com/openai-agents-sdk.md) | Open-source Python framework for building agents with tools, handoffs, guardrails, sandboxes and tracing. | MCP servers are a first-class external tool source, alongside human-in-the-loop approval. | — | Yes |
| [Pydantic AI](https://theagentsindex.com/pydantic-ai.md) | Typed Python agent framework: any model, any surface, with a harness for long-running coding and research agents. | MCP servers are a first-class tool source for its agents. | — | Yes |
| [AutoGen](https://theagentsindex.com/microsoft-autogen.md) | Microsoft's open-source Python framework for building single- and multi-agent applications; now in maintenance mode. | The official `autogen-ext` package is an MCP client over STDIO, SSE or Streamable HTTP. | — | Yes |
| [Mastra](https://theagentsindex.com/mastra.md) | Open-source TypeScript framework for building AI agents, with a hosted platform for running and observing them. | Both directions via dedicated packages: `MCPClient` connects out, `MCPServer` exposes Mastra's own tools to any MCP client. | — | Yes |
| [Letta](https://theagentsindex.com/letta.md) | Stateful agent harness, cloud and SDK for agents with persistent, git-versioned memory. | Acts as an MCP host, connecting agents to external servers' tools, resources and prompts. | — | Yes |
| [Dify](https://theagentsindex.com/dify.md) | Open-source platform for building agentic workflows, AI agents and RAG knowledge pipelines. | Any Dify app can be published as an MCP-compatible tool. | — | Yes |
| [Relevance AI](https://theagentsindex.com/relevance-ai.md) | Hosted low-code platform for building, evaluating and running teams of business AI agents. | Agent actions inside the platform are directly extendable by connecting them to external MCP servers. | — | Yes |
| [n8n](https://theagentsindex.com/n8n.md) | Workflow automation platform for building and running AI agents on a visual canvas, with code where needed. | Both directions: a dedicated MCP Server Trigger node and an MCP Client node. | — | Yes |
| [Salesforce Agentforce](https://theagentsindex.com/salesforce-agentforce.md) | Salesforce's platform for building, deploying and monitoring AI agents on top of CRM data. | A native MCP client connects to any MCP-compliant server, plus verified AgentExchange partner servers. | — | Yes |
| [Perplexity](https://theagentsindex.com/perplexity.md) | AI answer engine with an agentic Computer, the Comet browser, and a developer API for web-grounded search. | Both directions: an official MCP server, plus local MCP client support. | — | Yes |
| [Manus](https://theagentsindex.com/manus.md) | General-purpose AI agent that plans and executes tasks in its own cloud computer. | Prebuilt MCP connectors (client) to Notion, Linear, Gmail, Stripe and more. | — | Yes |
| [GPT Researcher](https://theagentsindex.com/gpt-researcher.md) | Open-source autonomous research agent that writes cited long-form reports from web and local sources. | An MCP server is available via a companion project (`gptr-mcp`), not bundled in the main repo. | — | Yes |
| [Elicit](https://theagentsindex.com/elicit.md) | Research agent that searches 138M+ papers and runs systematic reviews with sentence-level citations. | An official MCP server exposes its own search and extraction API. | — | Yes |
| [Vapi](https://theagentsindex.com/vapi.md) | Developer platform for building, testing and deploying voice AI agents that make and receive phone calls. | Custom tools are reachable through both native function calling and Model Context Protocol servers. | — | Yes |
| [ElevenLabs Agents](https://theagentsindex.com/elevenlabs-conversational-ai.md) | Hosted platform for building voice and chat agents, with telephony, SDKs and a choice of LLM. | Both directions: connects to external MCP servers and ships its own official server. | — | Yes |
| [Retell AI](https://theagentsindex.com/retell-ai.md) | Platform for building, testing and deploying AI voice and chat agents on phone, web and SMS, billed per minute. | Both directions: an MCP server for managing agents, plus mid-call client tool calls. | — | Yes |
| [LiveKit Agents](https://theagentsindex.com/livekit-agents.md) | Open-source Python and Node.js framework for voice AI agents, with LiveKit Cloud hosting, telephony and observability. | First-class client support, documented for both HTTP and stdio transports. | — | Yes |
| [PolyAI](https://theagentsindex.com/polyai.md) | Enterprise platform for building voice and chat agents that resolve customer calls, priced per minute. | Runs its own MCP server abstracting storage and data access, on top of broader MCP-style integrations across the platform. | — | Yes |
| [Sierra](https://theagentsindex.com/sierra.md) | Conversational AI platform for building enterprise customer-service agents across voice, chat, email and ChatGPT. | Agents are natively publishable to ChatGPT Apps via MCP (server direction). | — | Yes |
| [Fin](https://theagentsindex.com/intercom-fin.md) | Customer service AI agent priced per resolved outcome, deployable on any helpdesk. | A fully documented API, data connectors and MCP support for deeper integration. | — | Yes |
| [Crescendo](https://theagentsindex.com/crescendo.md) | AI-native CX platform: agents resolving support across voice, chat, email and SMS, with a staffed team behind them. | Connects its Assistant to external MCP servers, and exposes its own MCP config service. | — | Yes |
| [Cresta](https://theagentsindex.com/cresta.md) | Enterprise contact-centre platform with autonomous voice and digital AI agents plus real-time human agent assist. | MCP for standardized tool access, alongside its existing API-based CRM/billing integrations. | — | Yes |
| [Ada](https://theagentsindex.com/ada.md) | Enterprise AI customer-service agents that resolve inquiries across voice, chat, email and social channels. | Ships its own MCP server, letting external AI assistants (Claude Desktop, ChatGPT) query its systems directly. | — | Yes |
| [11x](https://theagentsindex.com/11x.md) | AI SDR and inbound phone agent for outbound prospecting and real-time lead qualification. | 'Custom MCP Integration' is listed on the Enterprise pricing tier, the thinnest evidence here: no docs, no stated client/server direction. | — | Yes |
| [Glean](https://theagentsindex.com/glean.md) | Enterprise AI platform for building, governing and running agents on permission-aware company context. | Ships MCP in both directions today: an MCP server exposing Glean's own search, chat and knowledge graph to external tools (Cursor, VS Code, Claude Desktop, ChatGPT), and an MCP host connecting 17+ preloaded remote servers. | — | Yes |
| [Zed](https://theagentsindex.com/zed.md) | Rust code editor with an Agent Panel running Zed's own agent, ACP external agents, or terminal agent CLIs. | MCP servers extend the agent's available context and tools: a client integration listed alongside its ACP support for external coding agents like Claude Code and Copilot. | — | No |
| [Microsoft Agent Framework](https://theagentsindex.com/microsoft-agent-framework.md) | Open-source SDK for AI agents and multi-agent workflows across .NET, Python and Go. | Native Model Context Protocol tool support ships as a named feature alongside Agent2Agent (A2A) protocol support: one of the framework's two headline 2026 interoperability standards. | — | Yes |
| [Botsify](https://theagentsindex.com/botsify.md) | White-label platform for building prompt-based AI chat and voice agents that agencies resell under their own brand. | MCP, document and web-search grounding ships as a named feature on both the Done-for-You and White Label plans, letting the agent reach tools and data beyond a static script. | — | Yes |
| [Google Agent Development Kit (ADK)](https://theagentsindex.com/google-adk.md) | Open-source framework for building, evaluating and deploying AI agents in Python, TypeScript, Go, Java or Kotlin. | Both directions: native Model Context Protocol tool support ships alongside Agent2Agent (A2A) protocol support, including a bundled MCP Toolbox for 30+ databases on the Go SDK. | — | Yes |
| [Claude Agent SDK](https://theagentsindex.com/claude-agent-sdk.md) | Python and TypeScript library that runs Claude Code's agent loop and tools inside your own application. | Both directions: external MCP servers connect the same way Claude Code does, and the `@tool` decorator mounts custom functions as an in-process MCP server with no separate subprocess to manage. | — | Yes |
| [Cartesia](https://theagentsindex.com/cartesia.md) | Real-time speech models and a managed runtime for production phone and web voice agents. | Publishes its own MCP server (cartesia-mcp) so clients like Cursor and Claude Desktop can call its TTS, STT and voice-management APIs directly: a server-side implementation rather than a client one. | — | Yes |
| [Webhound](https://theagentsindex.com/webhound.md) | Pay-as-you-go web research agent: set a dollar budget, get cited reports, datasets and claim traces. | Ships a native MCP server, so an agent inside Claude, Codex or Cursor can trigger a research run and read its Report or Dataset output directly. | — | Yes |
| [Agno](https://theagentsindex.com/agno.md) | Python SDK, open-source AgentOS runtime and hosted control plane for building and operating agents. | Native support in both directions: consumes external MCP servers as tools, and can expose an Agno agent as an MCP server itself. | — | Yes |
| [Botpress](https://theagentsindex.com/botpress.md) | AI agent platform with a visual studio, a code-first TypeScript ADK and a helpdesk, priced per conversation. | Bi-directional MCP support: consumes external MCP servers and exposes its own capabilities as one, alongside native WhatsApp/Slack/Teams/voice deployment. | — | Yes |
| [OpenAI Realtime API](https://theagentsindex.com/openai-realtime-api.md) | OpenAI's speech-to-speech API for low-latency voice agents over WebRTC, WebSocket or SIP. | Connects to remote MCP servers for external tool access, shipped alongside native SIP telephony and image input at general availability. | — | Yes |
| [Augment Code](https://theagentsindex.com/augment-code.md) | Cosmos platform running fleets of coding agents across the SDLC, plus Auggie CLI and IDE extensions. | MCP is named directly alongside its GitHub, GitLab, Slack, Linear and Jira triggers in its own already-published integrations field: a client integration, not a generic API mention. | — | Yes |
| [Gemini Code Assist](https://theagentsindex.com/gemini-code-assist.md) | Google's per-seat AI coding assistant and agent for IDEs, the terminal and GitHub pull requests. | Agent Mode (Enterprise, preview) is documented as completing multi-step tasks using system tools and Model Context Protocol (MCP) servers: a client integration named directly in its own already-published feature description, not a generic API mention. | — | No |
| [SalesTouch](https://theagentsindex.com/salestouch.md) | Remote MCP server giving AI agents typed LinkedIn tools with queues, limits, approvals and activity logs. | A remote MCP server is the product's only interface: its own listing states it exposes LinkedIn actions to Claude, Codex, Hermes and OpenClaw through MCP instead of a dashboard or browser extension. | — | Yes |
| [goose](https://theagentsindex.com/goose.md) | Open-source local AI agent — desktop app, CLI and ACP server — running on the LLM provider you bring. | An MCP client and host: connects to 70+ Model Context Protocol extensions spanning databases, browsers, GitHub and Google Drive, and MCP Apps let an extension render interactive UI inside Goose Desktop instead of plain text. | — | Yes |
| [TinyFish](https://theagentsindex.com/tinyfish.md) | Web infrastructure for AI agents: search, page fetch, cloud browsers and goal-driven web automation. | Ships as an MCP server: listed on the MCP registry and built for MCP clients like Claude and Cursor to call directly into its Search, Fetch, Agent and Browser APIs, alongside community nodes for n8n and Dify. | — | Yes |
| [MindStudio](https://theagentsindex.com/mindstudio.md) | No-code visual builder for AI agent workflows, published as web apps, APIs, webhooks or MCP servers. | Agents deploy as an MCP server, one of several publish targets alongside a REST API and Node.js package. | — | Yes |
| [Sim](https://theagentsindex.com/sim.md) | Open-source AI workspace for building, deploying and running agents across 1,000+ integrations. | MCP support is a named platform feature: workflows connect directly to Model Context Protocol tools and servers. | — | Yes |
| [Undermind](https://theagentsindex.com/undermind.md) | AI literature research agent that searches, reads and cites the scientific record. | Free tier at $0 is documented as 'Free at $0 with MCP,' naming MCP support directly on every pricing tier: the only entry here where the protocol is named on the pricing page itself rather than in the docs or integrations. | — | Yes |

## How this list is built

A tool earns a place here only when its own docs, GitHub repo, blog or changelog make an explicit MCP claim, as a client, a server, or both, not a generic "integrations" or "API" mention. Either side of the protocol counts: a tool other agents can call into qualifies exactly as one that calls out.

Two things do not. A machine-readable discovery endpoint such as /api/mcp indexes a vendor's own catalogue and is neither a client nor a server, which is why AgentsKit is recorded as an absence. And a vendor's own blog describing MCP as planned or in development is not a shipped capability, which is why Moveworks is.

Most entries here were confirmed in our original MCP census (see the full article for the per-tool sourcing). Those added to the index afterward (Warp, CodeRabbit, Salesforce Agentforce, Antigravity CLI, PolyAI, Mastra, Letta, Factory and Glean) were independently re-verified against primary vendor docs on 2026-07-25, the same pass that checked Unify and Moveworks and confirmed neither supports MCP yet.

A later group covers Zed, Microsoft Agent Framework, Botsify, Google Agent Development Kit, Claude Agent SDK, Cartesia, Webhound, Agno, Botpress, OpenAI Realtime API, Augment Code and Gemini Code Assist. These were confirmed from each tool's own already-Published, sourced listing fields (integrations and key-features) rather than a fresh vendor-docs pass, each naming MCP explicitly as a documented integration. SalesTouch was added on 2026-08-17, on a listing that describes a remote MCP server as the product's only interface.

Goose and TinyFish were added on 2026-08-18, the day after each was published. Goose is a documented MCP client and host connecting to 70+ extensions, with MCP Apps letting an extension render interactive UI inside Goose Desktop; TinyFish is an MCP server listed on the MCP registry and built for MCP clients such as Claude and Cursor to call into directly.

These facts are normally recorded on the tool's own listing as well, not stated here alone. But that is a property of our corpus on a given day rather than a guarantee: a 2026-08-16 research pass briefly dropped the MCP line from Devin, LlamaIndex and Replit Agent's own integrations fields; each was restored within a day, and all three name MCP directly again as of this check. MCP adoption is moving fast in 2026; treat this as a snapshot, not a ceiling, and check a specific vendor's current docs before relying on it for a real integration.

MindStudio and Sim were both added on 2026-08-18, the same day each was published. MindStudio names an MCP server as one of several agent deploy targets, alongside a REST API and Node.js package; Sim lists MCP support as a named platform feature, connecting workflows directly to Model Context Protocol tools and servers.

Undermind was added on 2026-08-25, the first re-review to add a tool previously recorded here as a confirmed absence. Its own pricing page names MCP support directly on every tier: Free ($0 "with MCP"), Pro ($16/mo billed annually), Team ($15/person/mo) and Enterprise (custom). That is the bar this page was written to detect: an explicit MCP claim in the listing, even when that claim lives on the pricing page rather than the docs or integrations.