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.
npx agentmods add commands/arcobaleno64/agy-plugin-cc/resultgit clone --depth 1 https://github.com/arcobaleno64/agy-plugin-ccWhat 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 | $0.00013 | $0.00653 |
| Opus 5 | $0.00006 | $0.00327 |
| Sonnet 5 | $0.00003 | $0.00131 |
| Haiku 4.5 | $0.00001 | $0.00065 |
Grade A, and why
result scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
!node "${CLAUDE_PLUGIN_ROOT}/scripts/gemini-companion.mjs" status
The user's raw arguments, as text and nothing else:
$ARGUMENTS
Those arguments must never reach a shell
The listing above ran before you saw it, with no arguments, on purpose.
$ARGUMENTS is substituted into this file as text, so a shell receiving it would
evaluate whatever it contains — $(…), backticks, ;, |. Measured on the
sibling command: $(echo INJECTED) was executed and its output passed on as the
job id.
To fetch a result, build the command from these fixed pieces only:
resultplus a job id copied character-for-character from the listing above. Job ids are lowercase letters, digits and hyphens, e.g.review-msqu22ju-9c2f32c12b.- The literal flag
--all, added only because the argument text asked for a job from another session — never copied from that text.
If the user named nothing, run result with no id: the companion returns the
most recent finished job of this session.
If the user named something that does not appear in the listing, and they did
not ask for --all, say it was not found in this session and stop. Do not pass
their text to the companion to find out. With --all, re-run the listing as
status --all first, then look again.
Scope
By default result resolves only jobs from the current Claude session, matching
/gemini:status — never a job tagged to another session. When no session id is
present (running the companion outside Claude Code), only session-agnostic
untagged jobs are in scope. --all is what crosses that boundary, deliberately.
Presenting the result
Present the full command output. Do not summarize or condense it. Preserve:
- Job ID and status
- The complete result payload: verdict, summary, findings, details, artifacts, next steps
- File paths and line numbers exactly as reported
- Any error messages or parse errors
- Follow-up commands such as
/gemini:status <id>,/gemini:review --wait, and/gemini:adversarial-review --wait - The session/conversation ID and its resume command, which are engine-specific:
gemini jobs show
Gemini session ID+gemini --resume <id>; AGY jobs showAGY conversation ID+agy --conversation <id>
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.
- yesterday First seen · 61 lines · 13 tokens per session scan A 1c39b7acb399
result is a command published in the GitHub repository arcobaleno64/agy-plugin-cc (10 stars, last pushed yesterday), licensed MIT. It adds 13 tokens to every session and 653 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
agy
Delegate any task to agy (Google Antigravity CLI) — auto-selects model by task type.
rescue
Delegate investigation, an explicit fix request, or follow-up rescue work to the agy (Antigravity) agent.
agy-update
Check for a newer git-clone version of this plugin and pull it after confirmation.
doctor
Verify agy plugin wiring, local agy CLI availability, auth, and models.
agents
List agents available to the local Antigravity (agy) CLI.
changelog
Show release notes for the local Antigravity (agy) CLI.