About Commercializer
Commercializer connects an AI assistant to governed registers and company workspaces for medical device and life-science commercial strategy. It cites records, retains selected findings, and supports guided workflows for decisions and planning.
Key features
- 01Search governed registers
- 02Cite source records
- 03Keep company findings
- 04Run guided workflows
- 05Ask the Commercializer agent
Use cases
- 01Assess regulatory pathways for a device
- 02Prepare a market-entry plan
- 03Compare competitor registrations and recalls
- 04Create submission checklists and company briefs
MCP support
Exposes Commercializer workspace data and workflows through its Streamable HTTP MCP server at https://commercializer.ai/mcp. Connection uses OAuth 2.1 or an API key, and actions are recorded in the workspace.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://commercializer.ai/mcpclaude mcp add --transport http ai-commercializer-commercializer https://commercializer.ai/mcp{"mcpServers":{"ai-commercializer-commercializer":{"type":"http","url":"https://commercializer.ai/mcp"}}}Open in Cursor{"servers":{"ai-commercializer-commercializer":{"type":"http","url":"https://commercializer.ai/mcp"}}}Open in VS Code{"mcpServers":{"ai-commercializer-commercializer":{"type":"http","url":"https://commercializer.ai/mcp"}}}{"mcpServers":{"ai-commercializer-commercializer":{"serverUrl":"https://commercializer.ai/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
Details
From the project README
Commercializer
Commercial strategy for medical device and life-science companies, read from the governed registers.
Bring it into the AI assistant you already use.
Connected, your assistant works on your company the way the Commercializer app does. It reads the governed registers (FDA clearances and approvals, EU and other registrations, recalls and adverse event reports, registered trials, patents, payments to clinicians, public tenders and company filings), cites every record it uses, keeps the findings worth remembering on your company's profile, and runs workflows that deliver a pathway decision, a submission checklist, a market entry plan, a competitor footprint or a company brief, asking you at each decision.
Everything it does happens as you, in the workspace you choose, and is listed on the MCP page of your workspace under its name, where you can disconnect it at any time.
- The server:
https://commercializer.ai/mcp(Streamable HTTP; sign in with OAuth 2.1, or use an API key). - This repository:
github.com/Commercializer-ai/commercializer-mcp, the plugin and a guide for every assistant. - The page: commercializer.ai/product/assistant.
- Let the assistant set itself up: give it
https://commercializer.ai/mcp/llms.txt. It says how to connect from each client, how to sign in, how to work the workspace and every tool.
Connect your assistant
| Assistant | Fastest way | Full guide |
|---|---|---|
| ChatGPT | Plugins → Add → Add custom MCP server, URL https://commercializer.ai/mcp, OAuth | ChatGPT |
| Claude (claude.ai, desktop, mobile) | Customize → Connectors → Add → Add custom connector | Claude |
| Claude Code | /plugin marketplace add Commercializer-ai/commercializer-mcp | Claude Code |
| GitHub Copilot (VS Code, Copilot CLI, JetBrains, Visual Studio) | Add to VS Code or copilot plugin marketplace add Commercializer-ai/commercializer-mcp | GitHub Copilot |
| Copilot coding agent (on github.com) | Repository settings → Copilot → MCP servers, with an API key | Copilot coding agent |
| Microsoft Copilot Studio (agents for Microsoft 365 Copilot and Teams) | Tools → Add a tool → Model Context Protocol, OAuth dynamic discovery | Copilot Studio |
| Cursor (editor and CLI) | Add to Cursor, or cursor-agent mcp login commercializer | Cursor |
| Codex (CLI, IDE, app) | codex mcp add commercializer --url https://commercializer.ai/mcp | Codex |
| Antigravity (and Gemini CLI) | serverUrl in mcp_config.json | Antigravity |
| Devin Desktop, Windsurf, Kiro, Zed, Goose and others | the server URL, or mcp-remote for a stdio-only client | Other clients |
| A script or your own code | an API key as Authorization: Bearer | API keys |
| Any assistant, on its own | give it https://commercializer.ai/mcp/llms.txt | For assistants |
Signed in to Commercializer, the MCP page of your workspace walks the same steps for the assistant you pick and makes an API key when you need one.
The plugin
This repository is also a plugin: the server, a skill that teaches the assistant how to work the workspace (read before stating a figure, cite every read, keep findings, use the tools, run workflows), and five skills you can call by name.
| Skill | What it does |
|---|---|
ask | Hands a question to Commercializer's agent; the answer comes back with every record cited. |
tool | Answers one question with a tool, every row with its sources. |
run | Starts a workflow on your company and follows it, answering each checkpoint with you. |
workspace | Your company and its competitors, what is kept, the watches running, the workflows your plan runs. |
setup | Connects and checks it works. |
It follows the open Agent Plugins format (plugin.json, mcp.json, skills/) that
ChatGPT, Codex, GitHub Copilot, VS Code and Cursor read, and carries a Claude Code manifest (.claude-plugin/) beside it.
Claude Code /plugin marketplace add Commercializer-ai/commercializer-mcp
/plugin install commercializer@commercializer
Copilot CLI copilot plugin marketplace add Commercializer-ai/commercializer-mcp
copilot plugin install commercializer@commercializer
VS Code Command Palette → Chat: Install Plugin From Source → https://github.com/Commercializer-ai/commercializer-mcp
Codex codex plugin marketplace add Commercializer-ai/commercializer-mcp
codex plugin add commercializer@commercializer
Cursor Cursor Settings → Plugins → https://github.com/Commercializer-ai/commercializer-mcp
It is also listed in the MCP Registry as ai.commercializer/commercializer.
What it can do
Every tool, with what it reads or changes. In short:
- Read the registers:
resolve,company_footprint,company_profile,device_trace,graph_compare,lookup,trials_find,recalls_find,document_search,document_get. Every answer starts with a read id you can cite. - Know your company:
search,fetch,list_profiles,get_profile,recall_findings,keep_finding,forget_finding. - Answer one question with a tool:
list_tools,run_tool,get_tool_run; 48 tools, from a US pathway and an FTO screen to open tenders, a trial landscape and who is investing in your space. - Run workflows and watches:
list_workflows,start_workflow,get_workflow_run,answer_checkpointand more; 31 workflows across regulatory, commercial, evidence, innovation, capital and programme work. - Ask the agent:
ask_agent,get_agent_answer.
Try:
- What clearances and recalls does our closest competitor hold, register by register?
- Which US pathway fits our device, and which predicates would we lean on?
- Run a market entry workflow for Germany and the UK on our lead product.
Privacy and control
- It works as you, in the one workspace you approved, and can do only what you can do there.
- Every read, finding and run it makes is marked with its name on the MCP page and on the profile.
- Disconnect an assistant or revoke a key on the MCP page; it stops at once.
- Workflow runs and research count against your workspace's plan, as the app's own do.
More in Security and data. Problems: Troubleshooting.
Support
commercializer.ai/contact · [email protected] · Privacy · Terms
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.