Gemini CLI
Google Gemini CLI: open-source command-line AI agent with file, shell, and web access. Uses the Gemini API with a generous free tier.
Agent Passport
APAI.agent-passport.v0.1Agent Passport
local-toolunverified v0.1
Agent: Gemini CLIVendor: GoogleVersion: varies (rolling release)
Install modes:local-tool
↓
Can read
- ·Files and directories in current working directory
- ·Shell command output
- ·Web pages (via built-in web fetch tool)
↑
Can write
- ·Files in current working directory
- ·Shell-writable destinations
↔
Can access
- ·Gemini API (for model inference)
- ·Public web URLs (via web fetch tool)
- ·MCP servers when configured
- ·Google Search (via built-in grounding tool)
$
Can spend
- ·Gemini API tokens per session (free tier: 1,000 requests/day with Gemini 2.5 Pro as of May 2026)
○
Can expose
- ·File contents and shell output sent to Gemini API for inference
- ·Web fetch contents included in API context
✓
Will request approval for
- ·Destructive shell commands (prompted before execution)
- ·File writes outside the working directory
v0.1 note: This passport is descriptive, not enforceable. Runtime enforcement is the Policy Pack + Gateway story (Phase 4+). Fields sourced from publicly documented agent behavior as of May 2026.
Compatible packages (v0.1)
APAI packages that declare support for local-tool environments compatible with this agent. v0.1 match is by platform overlap; full capability-field matching ships with Phase 4.