IDE Integrations

Add BioContext7 to your AI coding assistant. Every client below supports the Model Context Protocol (MCP) and can connect via local stdio or remote HTTP.

Checking status…

Sends an MCP initialize request to https://biocontext7-api.fly.dev/mcp

Quick start

npx @biocontext7/mcp

Runs the BioContext7 MCP server locally via stdio. Point any MCP-compatible client at this command.

Must have

Primary AI coding assistants with full MCP support.

▶Claude CodeTier 1

One-click install

claude mcp add biocontext7 -- npx @biocontext7/mcp

Remote (HTTP)

{
  "mcpServers": {
    "biocontext7": {
      "type": "http",
      "url": "https://biocontext7-api.fly.dev/mcp"
    }
  }
}

Local (stdio)

{
  "mcpServers": {
    "biocontext7": {
      "command": "npx",
      "args": [
        "-y",
        "@biocontext7/mcp"
      ]
    }
  }
}

Add to ~/.claude/settings.json or project .mcp.json. The one-click command registers the server automatically.

▶CursorTier 1

Remote (HTTP)

{
  "mcpServers": {
    "biocontext7": {
      "url": "https://biocontext7-api.fly.dev/mcp"
    }
  }
}

Local (stdio)

{
  "mcpServers": {
    "biocontext7": {
      "command": "npx",
      "args": [
        "-y",
        "@biocontext7/mcp"
      ]
    }
  }
}

Add to .cursor/mcp.json in your project root. Cursor supports both remote and local MCP servers.

▶VS Code / GitHub CopilotTier 1

Remote (HTTP)

{
  "mcp": {
    "servers": {
      "biocontext7": {
        "type": "http",
        "url": "https://biocontext7-api.fly.dev/mcp"
      }
    }
  }
}

Local (stdio)

{
  "mcp": {
    "servers": {
      "biocontext7": {
        "type": "stdio",
        "command": "npx",
        "args": [
          "-y",
          "@biocontext7/mcp"
        ]
      }
    }
  }
}

Add to .vscode/settings.json. Requires the GitHub Copilot extension with MCP support enabled. You can also use the command palette: MCP: Add Server.

▶Claude DesktopTier 1

Remote (HTTP)

{
  "mcpServers": {
    "biocontext7": {
      "type": "http",
      "url": "https://biocontext7-api.fly.dev/mcp"
    }
  }
}

Local (stdio)

{
  "mcpServers": {
    "biocontext7": {
      "command": "npx",
      "args": [
        "-y",
        "@biocontext7/mcp"
      ]
    }
  }
}

Add to claude_desktop_config.json. On macOS: ~/Library/Application Support/Claude/claude_desktop_config.json. On Windows: %APPDATA%\Claude\claude_desktop_config.json.

▶WindsurfTier 1

Remote (HTTP)

{
  "mcpServers": {
    "biocontext7": {
      "serverUrl": "https://biocontext7-api.fly.dev/mcp"
    }
  }
}

Local (stdio)

{
  "mcpServers": {
    "biocontext7": {
      "command": "npx",
      "args": [
        "-y",
        "@biocontext7/mcp"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json. Open Windsurf Settings > MCP to manage servers via the UI.

Should have

Popular AI coding tools with growing MCP ecosystems.

▶JetBrains AI AssistantTier 2

Local (stdio)

{
  "mcpServers": {
    "biocontext7": {
      "command": "npx",
      "args": [
        "-y",
        "@biocontext7/mcp"
      ]
    }
  }
}

Go to Settings > Tools > AI Assistant > MCP Servers > Add. Paste the config above. Requires JetBrains IDE 2025.1+ with AI Assistant plugin.

▶ZedTier 2

Local (stdio)

{
  "context_servers": {
    "biocontext7": {
      "command": {
        "path": "npx",
        "args": [
          "-y",
          "@biocontext7/mcp"
        ]
      }
    }
  }
}

Add to ~/.config/zed/settings.json under the context_servers key.

▶ClineTier 2

Local (stdio)

{
  "mcpServers": {
    "biocontext7": {
      "command": "npx",
      "args": [
        "-y",
        "@biocontext7/mcp"
      ]
    }
  }
}

Open Cline settings in VS Code, navigate to MCP Servers, and add the config above. Cline uses the same MCP server format as Claude Desktop.

▶Amazon Q Developer CLITier 2

Local (stdio)

{
  "mcpServers": {
    "biocontext7": {
      "command": "npx",
      "args": [
        "-y",
        "@biocontext7/mcp"
      ]
    }
  }
}

Add to ~/.aws/amazonq/mcp.json. Requires Amazon Q Developer CLI with MCP support.

▶Gemini CLITier 2

Local (stdio)

{
  "mcpServers": {
    "biocontext7": {
      "command": "npx",
      "args": [
        "-y",
        "@biocontext7/mcp"
      ]
    }
  }
}

Add to ~/.gemini/settings.json or use gemini --mcp-server to register.

▶OpenAI Codex CLITier 2

Local (stdio)

{
  "mcpServers": {
    "biocontext7": {
      "command": "npx",
      "args": [
        "-y",
        "@biocontext7/mcp"
      ]
    }
  }
}

Add to ~/.codex/config.json. Requires the OpenAI Codex CLI with MCP enabled.

Nice to have

Additional clients and runtime options.

▶Roo CodeTier 3

Local (stdio)

{
  "mcpServers": {
    "biocontext7": {
      "command": "npx",
      "args": [
        "-y",
        "@biocontext7/mcp"
      ]
    }
  }
}

Open Roo Code settings, navigate to MCP Servers, and add the config. Compatible with the Claude Desktop MCP format.

▶Kilo CodeTier 3

Local (stdio)

{
  "mcpServers": {
    "biocontext7": {
      "command": "npx",
      "args": [
        "-y",
        "@biocontext7/mcp"
      ]
    }
  }
}

Add via Kilo Code MCP settings panel in VS Code.

▶Augment CodeTier 3

Local (stdio)

{
  "mcpServers": {
    "biocontext7": {
      "command": "npx",
      "args": [
        "-y",
        "@biocontext7/mcp"
      ]
    }
  }
}

Add via Augment Code MCP settings in your IDE extension.

▶AmpTier 3

Local (stdio)

{
  "mcpServers": {
    "biocontext7": {
      "command": "npx",
      "args": [
        "-y",
        "@biocontext7/mcp"
      ]
    }
  }
}

Add to ~/.amp/config.json under mcpServers.

▶WarpTier 3

Local (stdio)

{
  "mcpServers": {
    "biocontext7": {
      "command": "npx",
      "args": [
        "-y",
        "@biocontext7/mcp"
      ]
    }
  }
}

Add to Warp MCP configuration. Open Warp Settings > AI > MCP Servers.

▶LM StudioTier 3

Local (stdio)

{
  "mcpServers": {
    "biocontext7": {
      "command": "npx",
      "args": [
        "-y",
        "@biocontext7/mcp"
      ]
    }
  }
}

Add via LM Studio MCP integration panel. Requires LM Studio 0.3+ with MCP support.

▶DockerTier 3

Local (stdio)

docker run -i --rm ghcr.io/hordago-labs/biocontext7-mcp

Run the official BioContext7 MCP container. Use with any MCP client that supports stdio by setting the command to docker and the args to the run parameters above.

Transport modes

Transport	Command	Use case
stdio	npx @biocontext7/mcp	Local IDE usage (default)
SSE	npx @biocontext7/mcp --transport sse	Streaming clients
HTTP	npx @biocontext7/mcp --transport http	Remote / shared access

Environment variables

Variable	Default	Description
BIOCONTEXT7_API_URL	https://biocontext7-api.fly.dev	Backend API base URL
BIOCONTEXT7_API_KEY	—	API key for authentication
PORT	3100	Port for SSE / HTTP transports

Troubleshooting

Server not starting?

Make sure Node.js 18+ is installed and npx is available on your PATH. Run npx @biocontext7/mcp directly in your terminal to test.

Tools not appearing?

Restart your IDE after adding the MCP config. Some clients require a full restart rather than just a reload.

Connection timeout?

For remote HTTP configs, verify that https://biocontext7-api.fly.dev/mcp is reachable from your network. Switch to local stdio as a fallback.