MCP server to interact with LogSeq via its API.
The server implements multiple tools to interact with LogSeq:
- list_graphs: Lists all available graphs
- list_pages: Lists all pages in the current graph
- get_page_content: Return the content of a single page
- search: Search for content across all pages
- create_page: Create a new page
- update_page: Update content of an existing page
- delete_page: Delete a page
It's good to first instruct Claude to use LogSeq. Then it will always call the tool.
Example prompts:
- Get the contents of my latest meeting notes and summarize them
- Search for all pages where Project X is mentioned and explain the context
- Create a new page with today's meeting notes
- Update the project status page with the latest updates
You can configure the environment with LogSeq API settings in two ways:
- Add to server config (preferred)
{
"mcp-logseq": {
"command": "uvx",
"args": [
"mcp-logseq"
],
"env": {
"LOGSEQ_API_TOKEN": "<your_api_token_here>",
"LOGSEQ_API_URL": "http://localhost:12315"
}
}
}- Create a
.envfile in the working directory with the required variables:
LOGSEQ_API_TOKEN=your_token_here
LOGSEQ_API_URL=http://localhost:12315
To prepare the package for distribution:
- Sync dependencies and update lockfile:
uv syncSince MCP servers run over stdio, debugging can be challenging. For the best debugging experience, we recommend using the MCP Inspector.
You can launch the MCP Inspector via npm with:
npx @modelcontextprotocol/inspector uv --directory /path/to/mcp-logseq run mcp-logseq