Workflow
Wenlan Workflow for VS Code MCP Clients
Use Wenlan as a local memory server from VS Code surfaces that support MCP.
Article packet
Workflows
VS Code users with MCP-compatible AI tools
5 min read
01
Wenlan treats VS Code as another MCP client surface.
02
The daemon remains local and shared with other tools.
03
Use capture and recall around real project decisions.
01
Quick answer
If your VS Code AI surface supports MCP, connect it to Wenlan with the MCP-only setup path and use the same memory loop as other clients.
Wenlan centralizes the memory layer outside any one interface. VS Code can participate without making VS Code the only place where context exists.
02
When this problem appears
VS Code is often the center of coding work, but the AI context still fragments when terminal sessions, Claude Code runs, and editor chats each remember different things.
03
Connect the VS Code surface
Use the supported MCP client path rather than hand-editing memory.
- Run npx -y wenlan setup.
- From the workspace root, run ~/.wenlan/bin/wenlan mcp add vscode; it writes .vscode/mcp.json with servers.wenlan.
- Use ~/.wenlan/bin/wenlan mcp add vscode --dry-run to preview the workspace config before writing.
- In VS Code, confirm MCP server trust, use MCP: List Servers to start or restart the server, and enable/select Wenlan tools in Chat or Agent mode.
- Verify with doctor or capture/recall.
- Use spaces for separate project buckets.
VS Code workspace MCP setup
npx -y wenlan setup
~/.wenlan/bin/wenlan mcp add vscode
~/.wenlan/bin/wenlan mcp add vscode --dry-run04
What to check next
Client MCP support changes over time. Prefer Wenlan's generated config or dry-run output over copying stale snippets from old docs. VS Code Remote and Dev Containers run MCP servers where configured; install or configure Wenlan in the remote environment or handle localhost forwarding intentionally.
Try the local memory loop
Install Wenlan, connect your AI client, and verify that capture, recall, and handoff work on your machine.
FAQ