About Cosense MCP Server
Cosense MCP Server is an MCP server for Cosense projects that lets MCP clients read, search and edit Cosense pages. It is configured through environment variables for the project name and session cookie, suited to users who want Cosense content available in AI workflows.
Key features
- 01Retrieve a page by title
- 02List available pages in the project
- 03Search pages by query string
- 04Insert text after a specified line
Use cases
- 01Query Cosense page content from an AI client
- 02Search the project for pages containing a term
- 03Append notes to an existing page
MCP support
This project is an MCP server that exposes tools such as get_page, list_pages, search_pages and insert_lines to MCP clients.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMECosense MCP Server
A MCP server for Cosense.
Tools
The following tools are available for interacting with Cosense pages:
get_page: Retrieves a page with the specified titlelist_pages: Lists available pages in the resourcessearch_pages: Searches for pages containing the specified query stringinsert_lines: Inserts text after a specified line in a page
Installation
Prerequisites
Set the following environment variables:
COSENSE_PROJECT_NAME— your Cosense project name.COSENSE_SID— session cookie used for authentication. Required for writing pages and reading private pages. Treat this like a secret. See https://scrapbox.io/scrapboxlab/connect.sid for more details.
For VS Code Users
Use one of these one-click installation buttons:
Note: After clicking the button, replace
your_project_nameandyour_sidwith your actual values in the configuration. For manual setup, you can also edit.vscode/mcp.jsondirectly.
For Claude Desktop Users
Add one of the following configurations to your claude_desktop_config.json:
Using pnpm
{
"mcpServers": {
"cosense-mcp-server": {
"command": "pnpm",
"args": ["-s", "dlx", "@yosider/cosense-mcp-server"],
"env": {
"COSENSE_PROJECT_NAME": "your_project_name",
"COSENSE_SID": "your_sid"
}
}
}
}
Using yarn
{
"mcpServers": {
"cosense-mcp-server": {
"command": "yarn",
"args": ["dlx", "-q", "@yosider/cosense-mcp-server"],
"env": {
"COSENSE_PROJECT_NAME": "your_project_name",
"COSENSE_SID": "your_sid"
}
}
}
}
Using npx
This package depends on JSR-hosted packages. npx requires adding the JSR registry to ~/.npmrc first.
Step 1: Add JSR registry to ~/.npmrc
Linux/macOS:
echo "@jsr:registry=https://npm.jsr.io" >> ~/.npmrc
Windows (PowerShell):
echo "@jsr:registry=https://npm.jsr.io" >> $env:USERPROFILE\\.npmrc
Step 2: Add configuration
{
"mcpServers": {
"cosense-mcp-server": {
"command": "npx",
"args": ["-y", "@yosider/cosense-mcp-server"],
"env": {
"COSENSE_PROJECT_NAME": "your_project_name",
"COSENSE_SID": "your_sid"
}
}
}
}
Development
Running from Source
If you prefer to run the server from a local copy of this repository, build it first:
git clone https://github.com/yosider/cosense-mcp-server.git
cd cosense-mcp-server
pnpm install
pnpm run build
Then configure your MCP client to use the local build:
{
"mcpServers": {
"cosense-mcp-server": {
"command": "node",
"args": ["/path/to/cosense-mcp-server/build/index.js"],
"env": {
"COSENSE_PROJECT_NAME": "your_project_name",
"COSENSE_SID": "your_sid"
}
}
}
}
Debugging
Since MCP servers communicate over stdio, debugging can be challenging. We recommend using the MCP Inspector, which is available as a package script:
pnpm run inspect
The Inspector will provide a URL to access debugging tools in your browser.
Acknowledgments
This project is forked from funwarioisii/cosense-mcp-server.
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.