Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@BioMCP install the bioimage server and run a quick intensity analysis on my latest microscopy image"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
The open MCP platform for biology, bioinformatics, bioimaging, and scientific AI.
BioMCP follows a PowerMCP-style architecture: one installable distribution, one machine-readable registry, independent MCP servers, client configuration, diagnostics, tests, and a documentation site. The registry states what is installable, experimental, planned, or externally provided.
BioImage — local image inspection, intensity summaries, and thresholding primitives.
ImageJ / Fiji — bridge to an explicitly configured local ImageJ/Fiji installation.
LLM — OpenAI-compatible model access for agent workflows using environment-provided credentials.
BioNuclei — validated external scientific MCP adapter; BioMCP does not duplicate its scientific engine.
CellProfiler — planned adapter; not advertised as installable until an executable, tested integration exists.
Related MCP server: protein-design-mcp
pip install biomcp
biomcp list
biomcp install --servers bioimage --clients generic
biomcp doctorImage-analysis dependencies:
pip install 'biomcp[bioimage]'biomcp run bioimage
biomcp run imagej
biomcp run llmEach server is independently launchable over stdio, following the modular server-factory pattern used by PowerMCP.
biomcp install --servers bioimage,imagej,llm --clients claude-desktop
biomcp install --servers bioimage,llm --clients codexUse --dry-run to inspect changes before writing client configuration.
BIOMCP_IMAGEJ_EXECUTABLE points to ImageJ/Fiji.
BIOMCP_LLM_BASE_URL, BIOMCP_LLM_API_KEY, and BIOMCP_LLM_MODEL configure the LLM bridge. OPENAI_API_KEY is accepted as an API-key fallback.
BioMCP transports typed tool calls. It does not invent scientific measurements, benchmarks, or validation claims. Scientific computation remains under the authority of the underlying server.
The product site lives in docs/ and is published by GitHub Pages.
A new integration needs an executable server, registry metadata, tests, documentation, client configuration, and an explicit validation state before being promoted to installable or validated.
MIT.
This server cannot be installed
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
MCP server for progressive tool usage at any scale (see https://klavis.ai)
Personal assistant MCP server with search, execute, packages, jobs, secrets, and integrations.
Nifty's MCP server — exposes tasks, projects, messages, and files as tools for AI agents.
MCP server for secureFlows: token-free URL builders and integration-linting tools for AI agents.
