Wenlan Configuration

Use Settings in the desktop app or the CLI and plugin setup for the part you want to change. Client configuration belongs to the client; daemon settings affect the local service. Run doctor after setup changes.

Spaces separate work, personal, client, and project memory without requiring multiple Wenlan installs. Set a space for a shell session with WENLAN_SPACE, or define stable spaces in ~/.wenlan/spaces.toml.

List your Spaces first. If client-a does not exist, create it before launching Claude Code with that strict Space pin.

Each local client stores its own MCP configuration. Use wenlan connect and inspect its dry run before changing a client file.

For an example of using Wenlan memory in Claude Code after setup, follow the Claude Code memory guide.

The daemon normally listens on loopback. A non-loopback bind exposes its API beyond the local machine, so change it only when you understand the privacy implications. Set variables in the daemon process environment before it starts.

The app and daemon use local data locations; client configuration is stored separately by each client. Use the backup guide to identify the files to preserve.

Configure optional daemon-side models and keys in Wenlan. The connected AI client uses its own provider settings.