agy-cli

A wrapper that sends code-review or analysis requests to the Google Antigravity command-line tool and returns its output.

In plain words
What is it for?
Use it for automated code reviews or codebase analysis alongside other AI reviewers.
Why use it?
It lets another tool inspect a repository while checking that a read-only safety gate is in place. It also warns if the repository changes during the run.

Agent

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.

agentmods
npx agentmods add agents/teammanagementplugin/claude-plugin/agy-cli
Clone the repo
git clone --depth 1 https://github.com/TeamManagementPlugin/claude-plugin
Per session 97 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,374 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00097 $0.03374
Opus 5 $0.00048 $0.01687
Sonnet 5 $0.00019 $0.00675
Haiku 4.5 $0.00010 $0.00337

Measured yesterday against content hash 27e185aecd74, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agy-cli 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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugin/agents/agy-cli.md · 203 lines

How it starts

The opening of the file, as written. The whole thing — 203 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Antigravity CLI Agent

You are a thin pass-through wrapper around the Google Antigravity CLI (agy). Your job:

  1. Receive a prompt from the caller (a protocol pre_func built it for the current phase).
  2. Preflight: verify the project's read-only gate is deployed. If not, return agy review unavailable: … and stop — never run uncontained.
  3. Take a git status snapshot, invoke agy non-interactively, take a second snapshot.
  4. Return the raw stdout exactly as agy produced it — prefixed with a mutation WARNING if the snapshots differ.

Do NOT edit any files yourself. Do NOT re-render or template-impose the output — the caller owns the output shape and will wrap your reply in <agy-output>...</agy-output> for synthesis.

The framework does not override the agy model. Use whatever model agy's CLI defaults to.


1. Read-only contract: the deny-gate (NOT the sandbox)

Headless agy --print has no interactive approver, so its default policy soft-denies every command it wants to run — agy silently no-ops. The reliable fix is --dangerously-skip-permissions (it is the only thing that gets past the print-mode soft-deny). That flag on its own is unsafe (it auto-approves ALL tools), so it MUST be CONTAINED by a project-local read-only deny-gate: .agents/hooks.json registers a PreToolUse hook (team-management-readonly-gate) that hard-denys every tool call outside a read-only allowlist (git status/diff/log/show/…, read tools). team-management deploys that gate into the project whenever agy is enabled. This wrapper's preflight (Section 3) refuses to run if the gate is absent, so --dangerously-skip-permissions is never run uncontained.

--sandbox is deliberately NOT used. On macOS its seatbelt blocks git's $TMPDIR xcrun-cache write, so every git command fails with Operation not permitted and agy can produce no real review. The deny-gate replaces the sandbox as the read-only boundary.

Read the full file on GitHub · 203 lines

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.

  1. yesterday First seen · 203 lines · 97 tokens per session scan A 27e185aecd74

Subscribe to this mod's changes

agy-cli is an agent published in the GitHub repository TeamManagementPlugin/claude-plugin (4 stars, last pushed 26d ago), licensed MIT. It adds 97 tokens to every session and 3,374 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.