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/rescuegit clone --depth 1 https://github.com/arcobaleno64/agy-plugin-ccWrote 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/arcobaleno64/agy-plugin-cc/rescue)<a href="https://agentmods.dev/commands/arcobaleno64/agy-plugin-cc/rescue"><img src="https://agentmods.dev/badge/commands/arcobaleno64/agy-plugin-cc/rescue.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 | $0.00019 | $0.01295 |
| Opus 5 | $0.00010 | $0.00647 |
| Sonnet 5 | $0.00004 | $0.00259 |
| Haiku 4.5 | $0.00002 | $0.00129 |
Grade A, and why
rescue 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 3d 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.
This is a copy
86% identical to rescue — 29 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Invoke the gemini:gemini-rescue subagent via the Agent tool (subagent_type: "gemini:gemini-rescue"), forwarding the raw user request as the prompt.
gemini:gemini-rescue is a subagent, not a skill — do not call Skill(gemini:gemini-rescue) (no such skill) or Skill(gemini:rescue) (that re-enters this command and hangs the session). The command runs inline so the Agent tool stays in scope; forked general-purpose subagents do not expose it.
The final user-visible response must be Gemini's output verbatim.
Raw user request: $ARGUMENTS
The argument text above is passed to the subagent as a prompt, not to a shell,
and it must stay that way: $ARGUMENTS is substituted into this file as text,
so it must never reach a shell — a shell would evaluate $(…), backticks, ;
and | inside it. Measured on the job commands: $(echo INJECTED) was executed
before Node ever started. If this command ever needs to call the companion
directly, assemble that call from checked, literal flags rather than from this
text.
Execution mode:
- If the request includes
--background, run thegemini:gemini-rescuesubagent in the background. - If the request includes
--wait, run thegemini:gemini-rescuesubagent in the foreground. - If neither flag is present, default to foreground.
--backgroundand--waitare execution flags for Claude Code. Do not forward them totask, and do not treat them as part of the natural-language task text.--model,--effort, and--engineare runtime-selection flags. Preserve them for the forwardedtaskcall, but do not treat them as part of the natural-language task text.--timeout <seconds>is a runtime flag too. Preserve it the same way, and only if its value is entirely digits and between 30 and 3600.- If the request includes
--resume, do not ask whether to continue. The user already chose. - If the request includes
--fresh, do not ask whether to continue. The user already chose. - Otherwise, before starting Gemini, check for a resumable rescue thread from this Claude session by running:
node "${CLAUDE_PLUGIN_ROOT}/scripts/gemini-companion.mjs" task-resume-candidate --json
- If that helper reports
available: true, useAskUserQuestionexactly once to ask whether to continue the current Gemini thread or start a new one. - The two choices must be:
Continue current Gemini threadStart a new Gemini thread
- If the user is clearly giving a follow-up instruction such as "continue", "keep going", "resume", "apply the top fix", or "dig deeper", put
Continue current Gemini thread (Recommended)first. - Otherwise put
Start a new Gemini thread (Recommended)first. - If the user chooses continue, add
--resumebefore routing to the subagent. - A resumed turn that also edits files is refused on the gemini engine:
gemini --resumeaccepts onlylatest, so it would continue whatever gemini ran last, and a resumed conversation carries its own workspace for the edits to land in. If that refusal comes back, report it as-is — the user chooses between--fresh, resuming read-only, and--engine agy, which pins the conversation by id. Do not silently drop--resumeor--writeand rerun. - If the user chooses a new thread, add
--freshbefore routing to the subagent. - If the helper reports
available: false, do not ask. Route normally.
Operating rules:
- The subagent is a thin forwarder only. It should use one
Bashcall to invokenode "${CLAUDE_PLUGIN_ROOT}/scripts/gemini-companion.mjs" task ...and return that command's stdout as-is. - Return the Gemini companion stdout verbatim to the user.
- Do not paraphrase, summarize, rewrite, or add commentary before or after it.
- Do not ask the subagent to inspect files, monitor progress, poll
/gemini:status, fetch/gemini:result, call/gemini:cancel, summarize output, or do follow-up work of its own. - Leave
--effortunset unless the user explicitly asks for a specific reasoning effort. - Leave
--timeoutunset unless the user asks for more time, or a previous run of the same work timed out. It caps how long the turn may run and, with it, how much output the turn can produce. - Leave the model and engine unset unless the user explicitly asks for one.
- Leave
--resumeand--freshin the forwarded request. The subagent handles that routing when it builds thetaskcommand. - If the helper reports that Gemini/AGY is missing or unauthenticated, stop and tell the user to run
/gemini:setup. - If the user did not supply a request, ask what Gemini should investigate or fix.
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.
- 3d ago First seen · 63 lines · 19 tokens per session scan A d7a6e59e7720
rescue is a command published in the GitHub repository arcobaleno64/agy-plugin-cc (10 stars, last pushed 4d ago), licensed MIT. It adds 19 tokens to every session and 1,295 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. It is 86% identical to rescue, differing in 29 lines, and is treated as a copy.
Other commands, from other repositories
agy
Delegate any task to agy (Google Antigravity CLI) — auto-selects model by task type.
optimize
Command "optimize" from doccker/cc-use-exp, covering 第 0 步:范围选择, 统计文件数和行数, 第 1 步:项目画像, 扫描内容 and 输出格式.
review
Command "review" from doccker/cc-use-exp, covering 参数解析, 解析参数, 提取模式和目标, 如果指定了 commit,验证其有效性 and 模式 1:full(默认).
project-scan
扫描项目生成配置(CLAUDE.md/restart.sh/ignore/Docker).
design
技术设计(doc 文档框架 / checklist 质量检查).
requirement
需求分析(doc 文档框架 / interrogate 极刑审问).