Claude GeminiSAFE
Lightweight MCP integration bringing Google's Gemini AI capabilities to Claude Code with 1M+ token context window, smart model selection, and powerful code analysis tools
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
🚀 What's Coming Next: I'm building an AI Agent Feedback Loop System that enables intelligent collaboration between Claude Code and Gemini AI. This will create a continuous improvement cycle where both AI agents learn from each other's suggestions, creating smarter code analysis and more contextual development assistance. Starting with Claude Code, then expanding to other IDEs. Stay tuned!
A lightweight integration that brings Google's Gemini AI capabilities to Claude Code through MCP (Model Context Protocol)
This project connects Claude Code (your coding assistant) with Google's Gemini AI models. Think of it as adding a second AI expert to your development team - one that can read and understand massive amounts of code at once (1M+ tokens, which is like reading hundreds of code files simultaneously).
With this integration, you can ask Gemini questions about your code, get security reviews, performance suggestions, and architectural advice - all without leaving your coding environment. It automatically chooses the right AI model for each task: fast responses for quick questions, deeper analysis for complex problems.
Table of Contents
- Key Features
- How It Works
- Architecture Overview
- Key Benefits
- File Structure
- Project Structure
- Multi-Client Support
- Quick Start
- [Usage Exa
472e97d72e15OBSERVED · 2026-10-06Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add claude-gemini-mcp-slim --env GOOGLE_API_KEY=${GOOGLE_API_KEY} --env TEST_GOOGLE_API_KEY=${TEST_GOOGLE_API_KEY} -- npx -y [email protected]{
"mcpServers": {
"claude-gemini-mcp-slim": {
"command": "npx",
"args": [
"-y",
"[email protected]"
],
"env": {
"GOOGLE_API_KEY": "${GOOGLE_API_KEY}",
"TEST_GOOGLE_API_KEY": "${TEST_GOOGLE_API_KEY}"
}
}
}
}Exposed tools (3)
3 read · 0 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
gemini_analyze_code | read | Analyze specific code sections with focused insights |
gemini_codebase_analysis | read | Analyze entire directories using Gemini CLI |
gemini_quick_query | read | Ask Gemini CLI any development question for quick answers |
Trust audit
SAFEgrade B · trust 86/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | WARN |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (1 observation(s))
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (16)
("/etc/", "/proc/", "/sys/", "/dev/", "/var/", "/usr/", "/bin/", "/sbin/").bandit
.gitleaks.toml
.mypy.ini
.release-please-config.json
.release-please-manifest.json
"~/.ssh/id_rsa",
"/etc/passwd",
"../../../bin/sh",
with patch("gemini_mcp_server.GEMINI_MODELS", {"flash": "../../etc/passwd"}):"gemini_codebase_analysis", {"directory_path": "../../../etc/passwd"}"../../../etc/passwd",
"../../../home/user/.bashrc",
@commitlint/cli, @commitlint/config-conventional, husky
pytest, pytest-asyncio, pytest-cov, pytest-mock, pytest-xdist, pytest-timeout, pytest-benchmark, httpx
mcp, google-generativeai
Gates applied: no_behavioural_pass.
472e97d72e15full audit observations/trust-audit/mcp-server/cmdaltctr__claude-gemini.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-06 | 472e97d72e15 | SAFE | B | 86 | first audit |
Questions
What is the Claude Gemini MCP server?
Lightweight MCP integration bringing Google's Gemini AI capabilities to Claude Code with 1M+ token context window, smart model selection, and powerful code analysis tools
What tools does Claude Gemini expose?
3 in total: 3 read-only, 0 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is Claude Gemini safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (86/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.
What credentials does Claude Gemini need?
It reads GOOGLE_API_KEY and TEST_GOOGLE_API_KEY from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does Claude Gemini run?
It speaks stdio, so it runs as a local process your client starts. It is published on npm as claude-gemini-mcp-slim at 1.3.1.
How current is this page?
The grade is for one exact copy of the source (472e97d72e15), read on 2026-10-06. The repository is watched and re-audited when it changes.