Skip to main content
beton
MCP Integration

Beton + Gemini CLI

Google's AI agent meets open-source revenue intelligence

Google · AI Coding Agent

Get Started Free

WHAT GEMINI CLI CAN DO WITH BETON

Signal detection Signal detection with configurable confidence thresholds
Lead scoring Lead scoring based on behavioral product data
Signal routing Multi-destination signal routing
Account analysis Account health and engagement analysis
Signal filtering Historical signal querying and trend detection

EXAMPLE PROMPTS

PROMPT

> Use Beton to check for any PQL signals above 0.85 confidence

Queries Beton for high-confidence product-qualified lead signals

PROMPT

> Analyze the account data for enterprise-co.com through Beton

Runs account-level behavioral analysis and returns key insights

PROMPT

> List all expansion signals from the past month

Retrieves signal history filtered by type and time range

PROMPT

> Score and route the latest trial conversion signals to Pipedrive

Combines lead scoring with automatic CRM routing in one command

CONNECT BETON TO GEMINI CLI

Beton connects your data source (PostHog) to your destinations (Attio, HubSpot, Pipedrive, and more) via MCP. Gemini CLI acts as the interface — talk to your revenue data in natural language.

1

Install Gemini CLI: npm install -g @google/gemini-cli

2

Create or edit ~/.gemini/settings.json with Beton MCP config

3

Set your BETON_API_KEY in the environment variables

4

Run gemini — Beton tools are automatically discovered

~/.gemini/settings.json 
{
  "mcpServers": {
    "beton": {
      "command": "npx",
      "args": ["-y", "@getbeton/mcp-server"],
      "env": {
        "BETON_API_KEY": "your-api-key"
      }
    }
  }
}

Replace your-api-key with your Beton API key from your dashboard.

REVENUE OPS WORKFLOWS

Terminal-First Revenue Ops

Use Gemini CLI as your revenue ops copilot. Check signals, score leads, and route opportunities — all from the same terminal where you run your builds.

Automated Signal Triage

Ask Gemini to pull all new signals, categorize them by type (PQL, expansion, churn), and route each to the appropriate destination based on your rules.

Weekly Revenue Health Check

Run a weekly review by asking Gemini to analyze account activity, identify trending signals, and flag accounts that need attention.

FREQUENTLY ASKED QUESTIONS

Is Gemini CLI free to use with Beton?
Gemini CLI is open-source and free. Beton can be self-hosted for free as well. You only need API keys for both services.
Does this work with Google Cloud's Gemini API?
Gemini CLI uses Google's Gemini models. Beton connects as an MCP tool server — the AI model doesn't matter, only the MCP protocol support.
Can Gemini CLI access all of Beton's MCP tools?
Yes. All five Beton MCP tools (beton_check_signals, beton_score_leads, beton_route_signal, beton_analyze_account, beton_list_signals) are available through Gemini CLI's MCP integration.
How is this different from using Beton's API directly?
MCP adds a natural language layer. Instead of writing API calls, you describe what you want in plain English and Gemini figures out which Beton tools to call and how to chain them.

OTHER AI AGENT INTEGRATIONS

Beton + Claude Code

Let Claude Code handle your revenue ops while you ship product

Beton + Cursor

Revenue ops meets your AI-powered code editor

Beton + OpenClaw

The open agent framework, supercharged with revenue intelligence

Beton + Windsurf

Revenue intelligence inside the agentic IDE

Beton + Cline

Autonomous coding agent with revenue intelligence superpowers

View all integrations →

Start using Beton with Gemini CLI

Connect Beton to Gemini CLI via MCP and start detecting revenue signals in minutes. Free to self-host, no credit card required.

Get Started Free Book a Demo