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.
git clone --depth 1 https://github.com/AIgen-Solutions-s-r-l/agy-claude-pluginWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/aigen-solutions-s-r-l/agy-claude-plugin/agy-debug)<a href="https://agentmods.dev/commands/aigen-solutions-s-r-l/agy-claude-plugin/agy-debug"><img src="https://agentmods.dev/badge/commands/aigen-solutions-s-r-l/agy-claude-plugin/agy-debug.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00025 | $0.01754 |
| Opus 5 | $0.00013 | $0.00877 |
| Sonnet 5 | $0.00005 | $0.00351 |
| Haiku 4.5 | $0.00003 | $0.00175 |
Grade A, and why
agy-debug 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 8d ago.
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.
How it starts
The opening of the file, as written. The whole thing — 151 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Have Antigravity (Google's agy CLI) help debug an error as a
root-cause analyst, then relay its hypotheses. The user invoked:
/agy-debug $ARGUMENTS
$ARGUMENTS is the pasted error or stack trace followed by zero or more
file paths that are relevant to it. Separate the two: the trace is prose,
the trailing tokens that resolve to real files are the attachments.
Before you run
Print a short progress line so the wait is expected, e.g.:
Debugging with Antigravity (~30s-3m, agentic CLI)…
Gather the context
- Keep the error/stack trace text verbatim — it goes into the prompt.
- For each named file that exists, plan to attach its contents. Resolve paths
relative to the repo (use
git rev-parse --show-toplevelif you need the root). Usegit status --shortfor orientation on what's changed.
Bound it (hard cap — never blow past this)
The attached file content must be capped before it leaves the machine:
- Hard cap across all attachments: ~600 lines and ~24 KB, whichever is hit first.
- If a file (or the set) is over the cap, attach the most relevant region — the area around the line numbers in the stack trace — and note that the file was truncated to fit the cap (with the line range included).
- Never include secrets,
.envfiles, or any.gitignoredfiles. If a named path is ignored or looks like a secret, skip it and tell the user it was not attached.
Consent line (always visible)
Before sending, print exactly one line stating what leaves the machine, e.g.:
Attaching <N lines / M files> to Antigravity (Google) — code leaves the machine.
Fill in the real counts from the bounded attachments.
Transport via a dedicated temp dir (in-scope, prompt-free)
Copy only the bounded, consented file content into a fresh temp directory
and hand that to agy with --add-dir (so nothing else in the repo is
exposed). Do this with the in-scope mktemp/sed commands — not with
Claude's built-in Read/Write — so a non-bypass user gets no mid-command
prompt:
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.
- 8d ago First seen · 151 lines · 25 tokens per session scan A 3a22b98722e5
agy-debug is a command published in the GitHub repository AIgen-Solutions-s-r-l/agy-claude-plugin (1 stars, last pushed 3mo ago), licensed MIT. It adds 25 tokens to every session and 1,754 once invoked, about $0.0001 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.
Other commands, from other repositories
gemini-imagegen
Generate or edit an image via a Gemini (agy) session — Gemini's generateimage tool does what Claude cannot.
cloud-run-debug
Diagnose a failing Cloud Run service — Antigravity (agy/Gemini) digests the error logs cheaply, Claude infers the root cause and proposes a fix. Read-only by default; --apply writes the fix to a branch.
report
Generate a publication-grade HTML report from a markdown source using the TRAID Design System catalog (5 templates). Agy matches content to template, you confirm, agy produces final branded HTML.
notebook-query
Query the notebook knowledge base (SQLite) built by /agy:notebook — precise, grounded, cited. Ask in natural language ("sum the amounts by category", "which docs mention 'Acme Corp'", "build a project timeline") or pass raw SQL. Read-only. Use this when you need exact aggregates/lookups across a document corpus…
model
Show or set the agy (Antigravity / Gemini) model by writing /.gemini/antigravity-cli/settings.json — the reliable way to switch models (the unreliable --model flag silently falls back to the default). No agy call; takes effect immediately, no plugin reinstall.
record
Record a browser walkthrough of a URL using Antigravity (agy). Generates .webm video, screenshots, and a report. Auto-converts to MP4 if ffmpeg is available.