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/faeton/claude-grok-plugin/rescuegit clone --depth 1 https://github.com/faeton/claude-grok-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/faeton/claude-grok-plugin/rescue)<a href="https://agentmods.dev/commands/faeton/claude-grok-plugin/rescue"><img src="https://agentmods.dev/badge/commands/faeton/claude-grok-plugin/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 | $0.00020 | $0.00284 |
| Opus 5 | $0.00010 | $0.00142 |
| Sonnet 5 | $0.00004 | $0.00057 |
| Haiku 4.5 | $0.00002 | $0.00028 |
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 4d 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.
What it actually says
Forward the user's request to the grok-consult subagent (xAI Grok).
Raw slash-command arguments:
$ARGUMENTS
Routing rules:
- Invoke the
grok-consultsubagent via theAgenttool (subagent_type: "grok:grok-consult"), forwarding$ARGUMENTSas the prompt. - Do not answer the question or perform the review yourself — the subagent forwards it to Grok.
- Execution mode:
- If
$ARGUMENTSincludes--background, run the subagent in the background. - If it includes
--wait, run it in the foreground. - Otherwise default to foreground for a small, clearly bounded ask; background for a large or open-ended one.
- If
--file <path>and any--effort/--modelflags are for Grok — leave them in the forwarded text so the subagent passes them through.- Return the subagent's output verbatim. Do not paraphrase, summarize, or act on Grok's findings in this turn.
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.
- 4d ago First seen · 22 lines · 20 tokens per session scan A b8cc8979f954
rescue is a command published in the GitHub repository faeton/claude-grok-plugin (1 stars, last pushed 19d ago), licensed MIT. It adds 20 tokens to every session and 284 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
implement
Claude plans, Grok implements - write a concrete implementation plan, hand it to Grok as the worker, wait, then verify the result.
review
Run a Grok code review against local git state.
result
Show the stored final output for a finished Grok job in this repository.
transcript
Export the full Grok session transcript for a finished job (or any Grok session ID) as Markdown.
transfer
Transfer the current Claude Code session into a resumable Grok session.
setup
Check whether the local Grok Build CLI is ready and optionally toggle the stop-time review gate.