Power your AI workflows with DrugBank MCP
Ask complex questions and get decision-ready insights.
Use Cases
Move faster with the full biomedical picture
Prioritize the right assets and targets
Cut through fragmented data to evaluate scientific and commercial potential so you can move faster on partnership, licensing, and investment decisions.
Make sense of complex biology
Navigate drug, target, disease relationships and assess mechanisms of action and pathways to determine what's worth advancing and what isn't.
Stay ahead of the competitive landscape
Track pipeline activity, clinical progress, and competitor movement to spot opportunities and anticipate risks before they impact your strategy.
Getting Started
-
ChatGPT will prompt you to sign in using your DrugBank OS account.
-
Connect the DrugBank app and start chatting.
For best results, Thinking mode is recommended.
-
Integrate
In your AI tool or app, find "Add MCP Server" or a similar option under settings or integrations.
-
Configure
Paste the MCP URL into the server or address field.
https://mcp.drugbank.com/mcpRefer to your AI tool's documentation for specific setup instructions.
-
Authenticate
You will be prompted to log in with your DrugBank OS account.
A paid DrugBank OS account with MCP access enabled is required for activation. Contact sales
Frequently asked questions
What's an MCP?
Programmatic access lets AI tools like Claude and ChatGPT connect directly to DrugBank's structured knowledge graph using an open standard called Model Context Protocol (MCP). Instead of relying on general web results, your AI tool queries DrugBank's expert-curated data to answer your research questions.
Do I need a DrugBank account to use the MCP?
Yes. You'll need a paid DrugBank OS account with MCP access enabled to authenticate and access the MCP. To enable MCP access, contact your Customer Success Manager or Account Manager. If you don't have this account, contact sales to get started.
What kinds of questions can I ask?
Any research questions that you'd use DrugBank to find. Target evaluation, drug–target–disease relationships, clinical trial activity, competitive landscape analysis, and pipeline comparisons. For example: "What drugs target BTK and what are their trial phases and sponsors?" or "Give me a competitive snapshot of the PNH treatment landscape, including approved drugs, clinical trials, and drug targets."
How is this different from using an AI tool directly without DrugBank's MCP?
Without DrugBank connected, AI tools draw from general training data and the open internet, which can produce surface-level answers, hallucinations, or unverifiable claims. With DrugBank connected, responses are grounded in our traceable, continuously updated knowledgebase of deeply connected, expert-curated data.