rescue

Invoke the stereo:codex-rescue subagent via the Agent tool (subagent_type: "stereo:codex-rescue"), forwarding the raw user request as the prompt. stereo:codex-rescue is a subagent, not a skill — do not call Skill(stereo:codex-rescue) (no such skill) or Skill(stereo: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 Codex's output verbatim.

Raw user request: $ARGUMENTS

Execution mode:

• The Agent invocation and the subagent's single Bash call are always foreground, so there is always stdout to return verbatim.
• --background is forwarded to the companion task call as --background; its stdout is the launch record naming the job id and /stereo:status <jobId>.
• --wait is never forwarded; it selects a foreground companion task call.
• If neither flag is present, let the subagent choose whether to add task --background from the request's size and complexity.
• Do not treat --background or --wait as part of the natural-language task text.
• --model and --effort are runtime-selection flags. Preserve them for the forwarded task call, but do not treat them as part of the natural-language task text. If --model starts with claude:, stop before invoking the subagent and explain that /stereo:rescue is a Codex bridge: stereo:codex-rescue only forwards to the companion task runtime. For Claude-routed work, ask this session directly; use /stereo:quick --implementer claude:<alias> or /stereo:implement --implementer claude:<alias> for a contained Claude implementer; or use /stereo:adversarial-review --model claude:<alias> for a Claude review.
• 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 Codex, check for a resumable rescue thread from this Claude session by running:

node "${CLAUDE_PLUGIN_ROOT}/scripts/codex-companion.ts" task-resume-candidate --json

• If that helper reports available: true, use AskUserQuestion exactly once to ask whether to continue the current Codex thread or start a new one.
• The two choices must be:
  • Continue current Codex thread
  • Start a new Codex 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 Codex thread (Recommended) first.
• Otherwise put Start a new Codex thread (Recommended) first.
• If the user chooses continue, add --resume before routing to the subagent.
• If the user chooses a new thread, add --fresh before 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 Bash call to invoke node "${CLAUDE_PLUGIN_ROOT}/scripts/codex-companion.ts" task ... and return that command's stdout as-is.
• Return the Codex 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 /stereo:status, fetch /stereo:result, call /stereo:cancel, summarize output, or do follow-up work of its own.
• Model, effort, --resume, and --fresh handling belongs to the subagent's codex-cli-runtime skill: leave those flags in the forwarded request untouched and never treat them as task text. The runtime rejects claude:* values, so a Claude selection must never be forwarded.
• If the helper reports that Codex is missing or unauthenticated, stop and tell the user to run /stereo:setup.
• If the user did not supply a request, ask what Codex should investigate or fix.