Server facts
openevidence-mcp
Project
- Author
- bakhtiersizhaev
- License
- Not specified
- Category
- Browser Automation
Tech
- Runtime / language
- Not specified
- Transports
- stdio
- Remote available
- No
- Authentication
- Not specified
Activity
- GitHub stars
- 45
- Last commit
- Data as of
Tags
- Playwright
- clinical decision support
- evidence based medicine
- healthcare ai
- mcp server
- medical research
Overview
Clinical evidence research in OpenEvidence can start from an MCP client through this unofficial integration, which uses Playwright browser-session authentication rather than an API key. Its tools cover checking account access, searching earlier work, retrieving an article, submitting a research question, and waiting for an answer to finish. Users can check authentication with oe_auth_status before accessing account content. For revisiting research, oe_history_list supports pagination and optional search, while oe_article_get retrieves an identified article with normalized question, answer, and status fields.
Submitting through oe_ask creates a question and article in the user's OpenEvidence account; it is intended for research rather than medical advice or diagnosis for an individual patient. Longer requests can return an article identifier without waiting, letting oe_article_wait handle completion separately when host timeouts are a concern. An existing article identifier can also preserve continuity for a genuine follow-up. History and article tools omit raw content by default to reduce exposure of private material; requesting raw data requires explicit user intent. The integration communicates over stdio and supports Linux, Windows, and macOS.
openevidence-mcp setup notes
- openevidence-mcp records stdio only; confirm a network transport upstream before planning remote access.
- openevidence-mcp has no recorded license; confirm reuse terms in its repository.