Add the KnowMan server
Open Settings | Tools | AI Assistant | Model Context Protocol (MCP), then click Add. You can also open MCP settings from chat by typing / and selecting Add Command. JetBrains documents both paths.
Select the JSON configuration connection method and enter the snippet above. JetBrains uses mcpServers with command and args for local servers, and supports launching an uninstalled server with npx. See the JetBrains configuration guide.
Sign in once with knowman login
The JSON snippet shows "KNOWMAN_API_KEY": "km_..." as a placeholder. Create a key in KnowMan under Settings > Connections > Connect an AI agent, or run npx -p @knowman/mcp knowman login once. The CLI and MCP server share the saved credential, so after login you can remove the env block. KnowMan MCP setup.
KnowMan’s open-source MCP client runs locally over stdio with npx -y @knowman/mcp and communicates with KnowMan’s HTTP API. KnowMan also offers a hosted remote MCP server at https://api.knowman.io/mcp. KnowMan MCP setup.
Apply and check the connection
Click OK, then Apply to start the configured server and establish a connection. Check the Status column for connection status. JetBrains setup steps.
After connection succeeds, select the icon in the Status column to review available tools. JetBrains shows tools there.
Reconnect or inspect logs
To reconnect, select the server and click Reconnect. JetBrains documents this action.
For server logs, open the mcp folder inside the IDE logs directory. JetBrains identifies this log location.
KnowMan tools in JetBrains AI Assistant
KnowMan exposes six tools: kb_search searches saved items, kb_ask answers questions with citations, and kb_get_item retrieves one item’s readable content. KnowMan MCP tool list.
kb_save saves a URL with an optional note, kb_recent lists recent items, and kb_memories lists saved claims, decisions, and playbooks. KnowMan MCP tool list.
If setup options are missing
The MCP Server plugin is bundled and enabled by default in JetBrains IDEs. If the relevant features are unavailable, check whether the plugin was disabled. JetBrains plugin note.
An organization administrator may centrally manage MCP servers and control whether users can add their own. JetBrains describes administrator controls.
Use the hosted KnowMan server
For JetBrains AI Assistant, in clients that support remote MCP servers, add the server URL https://api.knowman.io/mcp. Sign in on the KnowMan consent page and approve access, or use an API key sent as an Authorization: Bearer <key> header.
After connecting, ask the assistant to search or save, for example, “search my KnowMan for …” or “save this URL.” KnowMan’s MCP tools include search, ask, reading items, saving links, recent items, and memories.
Questions people ask
Does KnowMan use local or remote MCP in JetBrains?
KnowMan’s MCP server uses local stdio, launched as a subprocess. JetBrains AI Assistant supports local STDIO servers. JetBrains MCP documentation.
Do I need to keep the API key in the JSON config?
No. After knowman login, the CLI and MCP server share the saved credential, so the env block can be removed. KnowMan MCP setup.
How do I see which KnowMan tools are available?
After the connection succeeds, click the icon in the Status column to review the server’s available tools. JetBrains MCP documentation. KnowMan’s available tools include kb_search, kb_ask, and kb_get_item. KnowMan MCP tool list.
Where can I find MCP server logs?
Open the mcp folder inside the IDE logs directory. JetBrains MCP documentation. The folder contains logs for each configured MCP server. JetBrains MCP documentation.