Agent integration / MCP companies, products & suppliers

Model Context Protocol and similar tool-wiring layers that connect models to tools. Jobs: mcp servers; tool adapters; host-to-tool permission mapping.

What is Agent integration / MCP?

Model Context Protocol and similar tool-wiring layers that connect models to tools. Jobs: mcp servers; tool adapters; host-to-tool permission mapping.

What problems does it solve?

Each model host has a different tool interface, so agents cannot share connectors.

Typical business use cases

  • MCP servers
  • Tool adapters
  • Host-to-tool permission mapping

Important capabilities

  • Protocol support
  • Auth to tools
  • Server catalog

What buyers should evaluate

  • Which hosts are supported
  • Secret handling
  • Allow-list of servers

Risks and governance considerations

A connector that grants a model broader tool rights than the user has.

Procurement checklist

  • Server allow-list
  • Auth model
  • Audit of tool calls

Relevant AI Trustmark assurance

AI Trustmark independent findings appear only when an assessment or certificate exists. Category membership does not imply verification.

Methodology · How verification works

Companies and providers

Claimed suppliers appear first so buyers can start with listings the company has taken ownership of. Payment does not buy this order.

  • Arcade

    Arcade is the actions runtime for enterprise AI agents — one control point to enforce, execute, and govern every action, every time. Arcade is used for Agent integration / MCP work

  • Cloudflare

    Cloudflare publishes Cloudflare DLP, Cloudflare Agents SDK MCP Support, and Cloudflare Workers AI as named AI products. Cloudflare is used for Agent integration / MCP and AI agents

  • Composio

    Just-in-time tool calls, secure delegated auth, sandboxed environments, and parallel execution across 1,500+ apps. Composio is used for Agent integration / MCP work. Composio publi

  • Docker

    Docker publishes Docker MCP Toolkit as named AI products. Docker is used for Agent integration / MCP work. Docker publishes product information at docker.com. Docker is grouped wit

  • GitHub

    GitHub, a Microsoft company, sells Git hosting, Actions CI and Copilot AI coding assistance. Public pages cover cloud repositories and developer tools. GitHub publishes GitHub Copi

  • Pipedream

    Managed auth and 10,000+ tools across 3,000+ APIs for the agents you’re building — and a secure gateway for the AI tools your company uses. Pipedream publishes Pipedream MCP as nam

  • Portkey

    Democratize and productionize Gen AI across your entire org with Portkey. Portkey publishes Portkey AI Gateway and Portkey MCP Gateway as named AI products. Portkey is used for AI

  • Postman

    Accelerate API development with Postman's all-in-one platform. Streamline collaboration and simplify the API lifecycle for faster, better results. Postman publishes Postman MCP Ser

  • Smithery

    Connect agents to thousands of tools and services so they can act with more agency. Smithery is used for Agent integration / MCP work. Smithery publishes product information at smi

  • Zapier

    Build and scale AI workflows and agents across 9,000+ apps with Zapier—the most connected AI orchestration platform. Zapier publishes Zapier Agents and Zapier MCP as named AI produ

Products

Claimed products appear first. Ranking packs and payment do not change this list.

Related categories

Relevant procurement and assurance guides

Frequently asked questions

What is Agent integration / MCP?

Model Context Protocol and similar tool-wiring layers that connect models to tools. Jobs: mcp servers; tool adapters; host-to-tool permission mapping.

What should not be listed as Agent integration / MCP?

Products whose buyer job is finished agents, orchestration fabrics, or generic iPaaS without an agent protocol. Those belong on their own category page so search queries are not split.

Has AI Trustmark independently assessed every Agent integration / MCP supplier?

No. A category listing is descriptive. Independent assessment is shown only on company or product pages that carry Trustmark evidence.

What should buyers check when an AI agent can take actions?

Ask which tools, systems and write actions are allowed, whether those actions require human approval, and how they can be stopped. Category membership as an AI agent is a topic label, not permission to act in a buyer environment.

How is an AI agent different from workflow automation?

Workflow automation follows a designed path, even if a model fills a step. Agents choose steps toward a goal. Both can write to systems; the control model differs. Do not treat an RPA bot or a predefined workflow as an open-ended agent in procurement language.

Should agent actions require human approval?

Human approval is a control, not a default. Buyers should specify which action classes need approval, who can grant it, and whether the agent can proceed unattended after a session starts. Independent assessment records the approval model that was evidenced.

What evidence should buyers request from an AI-agent supplier?

Request the named product, permitted actions, logging, approval model, independent assessment depth if any, and the live certificate record rather than a badge image. Supplier-stated answers are not the same as documented, demonstrated or independently verified evidence.

How should prompt injection and tool-output attacks be controlled?

Agents that read untrusted content or tool output can be instructed to exfiltrate data or take writes. Buyers should ask what is treated as untrusted, whether tool output can change the plan, and what tests were run. Scanner marketing is not the same as independent testing.