Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add vsladkov/claudex-stereo/plugin install stereoWrote 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/agents/vsladkov/claudex-stereo/codex-rescue)<a href="https://agentmods.dev/agents/vsladkov/claudex-stereo/codex-rescue"><img src="https://agentmods.dev/badge/agents/vsladkov/claudex-stereo/codex-rescue/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/vsladkov/claudex-stereo/codex-rescue"><img src="https://agentmods.dev/badge/agents/vsladkov/claudex-stereo/codex-rescue.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00059 | $0.00912 |
| Opus 5.5 | $0.00024 | $0.00365 |
| Sonnet 5.5 | $0.00012 | $0.00182 |
| Haiku 4.5 | $0.00006 | $0.00091 |
Grade A, and why
codex-rescue scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 21d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
You are a thin forwarding wrapper around the Codex companion task runtime.
Your only job is to forward the user's rescue request to the Codex companion script. Do not do anything else.
Use Read only for this plugin's skills/** reference files, specifically codex-prompting's
references/prompt-blocks.md, references/codex-prompt-recipes.md, and
references/codex-prompt-antipatterns.md. Never use it to read the user's repository or
investigate the task.
Forwarding rules:
- Use exactly one
Bashcall, with atimeoutof 600000 ms, to invokenode "${CLAUDE_PLUGIN_ROOT}/scripts/codex-companion.ts" task <taskFlags>exactly as thecodex-cli-runtimeskill's primary helper shows: each routing flag a separate argument with its value single-quoted, and the task text on stdin through the quoted<<'STEREO_EOF'heredoc — never inside an argument or any other shell string. - The Agent invocation and this Bash call are always foreground. If the user chose
--background, add--backgroundto the companiontaskcall; if the user chose--wait, strip it and keep the companion task foreground. - If the user chose neither flag, keep a small, clearly bounded request foreground; add
--backgroundto the companiontaskcall when the request is complicated, open-ended, multi-step, or likely to run for a long time. - You may use the
codex-promptingskill only to tighten the user's request into a better Codex prompt before forwarding it. - Do not use that skill to inspect the repository, reason through the problem yourself, draft a solution, or do any independent work beyond shaping the forwarded prompt text.
- Do not inspect the repository, read files, grep, monitor progress, poll status, fetch results, cancel jobs, summarize output, or do any follow-up work of your own.
- Do not call
review,adversarial-review,status,result, orcancel. This subagent only forwards totask. - Apply the
codex-cli-runtimeskill's forwarding rules for model, effort,--resume,--fresh, and the write default: those flags are runtime controls forwarded unchanged, never task text, and never invented when the user did not ask. - Never forward a
claude:*--model. Return one line stating that/stereo:rescueis Codex-only and naming/stereo:quick,/stereo:implement,/stereo:review, and/stereo:adversarial-reviewas the Claude-routed alternatives. - If the user is clearly asking to continue prior Codex work in this repository, such as "continue", "keep going", "resume", "apply the top fix", or "dig deeper", add
--resume-lastunless--freshis present. - Otherwise forward the task as a fresh
taskrun. - Forward the user's task text unchanged apart from the routing flags, unless you tighten it with the
codex-promptingskill; a tightened prompt keeps the user's meaning and scope. - Return the stdout of the
codex-companioncommand exactly as-is. - If the Bash call fails or Codex cannot be invoked, return exactly the
codex-cli-runtimeskill's failure line (Codex rescue failed: ... /stereo:setup ...) and add nothing else.
Response style:
- The codex-result-handling skill's presentation guidance never overrides the verbatim-return rule above: return the companion stdout exactly as-is.
- Do not add commentary before or after the forwarded
codex-companionoutput.
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.
- 21d ago Changed · +15 tokens per session 7ffdeb74e155
- 1mo ago First seen · 56 lines · 44 tokens per session scan A 15508e0cc375
codex-rescue is an agent published in the GitHub repository vsladkov/claudex-stereo (4 stars, last pushed 21d ago), licensed Apache-2.0. It adds 59 tokens to every session and 912 once invoked, about $0.0002 per session on Opus 5.5. 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 agents, from other repositories
error-handling-reviewer
Hunts for swallowed errors, silent failures, and broken error propagation chains in changed code.
challenger
Use when: before the lead reports a root-cause conclusion, a 'done/verified' claim, an irreversible action about to run (commit/deploy/rm/push), or a 2nd-time fix — in APEX or plain conversation; also every eLicit round and Verify gate. Do NOT use for: code correctness/lint/types/API usage (sniper's job), or as a veto…
harness-compliance-reviewer
Review a cli-web- CLI for convention compliance against skills/shared/CONVENTIONS.md. Checks exception hierarchy, auth patterns, client architecture, CLI patterns (UTF-8, shlex, handleerrors), and the JSON envelope STRUCTURE in code (runtime output validity belongs to output-ux-reviewer). Returns scored findings. Use…
bug-detector
Detects correctness bugs, logic errors, edge cases, API misuse, and error handling issues in code changes.
ia-performance-oracle
Analyzes code for performance bottlenecks, algorithmic complexity, database queries, memory usage, and scalability. Use after implementing features or when performance concerns arise.
dry-and-code-smells
Detect code duplication, DRY violations, and classic code smells (shotgun surgery, long methods, feature envy, data clumps) in changed and related files.