Gemini CLI

Integrate Vectora with the Google Gemini CLI to use Vectora within Gemini workflows.

Quick Setup

1. Install Gemini CLI

npm install -g @google/generative-ai-cli

2. Configure Vectora as an MCP Server

In your .gemini/config.json:

{
  "mcpServers": {
    "vectora": {
      "command": "vectora",
      "args": ["mcp", "--stdio"],
      "env": {
        "VECTORA_NAMESPACE": "your-project"
      }
    }
  }
}

3. Use in Gemini Prompts

gemini chat --system "You have access to Vectora to search for code"

Use Cases

Case	Description
Code Search	Search for relevant files via Gemini + Vectora
Documentation	Generate documentation from indexed code
Analysis	Analyze codebase patterns

Troubleshooting

Gemini not seeing Vectora?

Verify that vectora mcp --stdio runs locally.
Check permissions in .gemini/config.json.
Restart the Gemini CLI.

Questions? GitHub Discussions

External Linking

Concept	Resource	Link
Gemini API	Google AI Studio Documentation	ai.google.dev/docs
MCP	Model Context Protocol Specification	modelcontextprotocol.io/specification
MCP Go SDK	Go SDK for MCP (mark3labs)	github.com/mark3labs/mcp-go

Part of the Vectora ecosystem · Open Source (MIT) · Contributors

Custom Agents