Workflow
Wenlan Workflow for Gemini CLI
Connect Gemini CLI as an MCP client and use Wenlan for local capture, recall, and handoff-style notes.
Article packet
Workflows
Gemini CLI users adding persistent local memory
5 min read
01
Gemini CLI can use the same Wenlan daemon through MCP.
02
Capture durable facts from command-line work.
03
Recall context from other Wenlan-connected clients.
01
Quick answer
Set up Wenlan, add Gemini CLI as an MCP client, verify with gemini mcp list or /mcp list, and ask Gemini to use the Wenlan MCP tools for context, capture, recall, and doctor checks.
Wenlan keeps those durable lessons in the same local store used by GUI and coding clients, so terminal work can feed later AI sessions.
02
When this problem appears
CLI sessions are especially easy to lose. A useful command, setup gotcha, or verified diagnosis disappears unless it is captured while the evidence is fresh.
03
Use Gemini CLI with Wenlan
Treat the CLI as another surface over the same memory layer.
- Run Wenlan setup before adding the client.
- Make sure Gemini CLI is installed and on PATH.
- Add Gemini CLI with ~/.wenlan/bin/wenlan mcp add gemini.
- Verify with gemini mcp list or /mcp list inside Gemini CLI.
- Ask Gemini to use Wenlan MCP tools such as mcp_wenlan_context or mcp_wenlan_capture.
- Capture verified setup and debugging lessons.
- Recall by command, error string, or project name.
- Capture a concise handoff-style note before ending a CLI session that future agents may continue.
Gemini CLI MCP setup
npx -y wenlan setup
~/.wenlan/bin/wenlan mcp add gemini
gemini mcp list04
What to check next
Avoid capturing raw terminal output. Store the conclusion and the command that proved it when that matters. Wenlan uses Gemini's user-scope MCP setup; use project-scope or manual Gemini config only for isolated experiments.
Try the local memory loop
Install Wenlan, connect your AI client, and verify that capture, recall, and handoff work on your machine.
FAQ