gemini

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Security

Grade A, and why

gemini scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

plugins/gemini/.mcp.json · 13 lines

What it actually says

{
  "gemini": {
    "command": "node",
    "args": [
      "-e",
      "process.env.GEMINI_MCP_STDIO='1';const p=require('node:path');const r=String(process.env.GEMINI_PLUGIN_ROOT);const root=p.isAbsolute(r)?r:process.cwd();import(require('node:url').pathToFileURL(p.join(root,'scripts','gemini-mcp.mjs')).href).catch(function(e){console.error(e);process.exit(1)})"
    ],
    "cwd": ".",
    "env": {
      "GEMINI_PLUGIN_ROOT": "${CLAUDE_PLUGIN_ROOT}"
    }
  }
}

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

Related

Other mcp servers, from other repositories