一键导入
codex-cli-runtime
Internal helper contract for calling the multi-cli-companion runtime from Claude Code
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Internal helper contract for calling the multi-cli-companion runtime from Claude Code
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Rewire which CLI handles which role in cc-multi-cli-plugin, OR diagnose/work around an upstream CLI quirk via env vars and config files. Use when the user asks to swap CLIs, change a subagent's target CLI, add or disable a subagent or command, restrict a CLI to read-only, hardcode a model, or change how a role frames its prompt — and also when a CLI is misbehaving (hangs, missing tools, broken release) and the user needs operator escape hatches like CURSOR_AGENT_PATH or AGY_CLI_PATH or OPENCODE_CLI_PATH, or per-CLI MCP config tuning. Works for any CLI in the marketplace — the four default CLIs (Codex, Cursor, Antigravity, OpenCode) and any additional CLIs the user added via the multi-cli-anything skill. Trigger phrases include "swap Codex and Cursor", "make Antigravity the researcher", "disable cursor-explore", "restrict Codex to read-only", "change which CLI handles implementation", "add /<cli>:<command>", "only install the plugins I need", "hardcode a model for <some-role>", "change the framing for <role>",
Add a new CLI provider to cc-multi-cli-plugin (beyond the built-in Codex/Cursor/Antigravity/OpenCode). Use when the user asks to integrate another AI CLI like Aider, Qwen, or any CLI that can be driven headlessly (a `-p`/print mode with JSON output, an app-server/HTTP mode, or any structured transport). Trigger phrases include "add Aider to the plugin", "integrate Qwen", "hook up my custom CLI", "support another model via its CLI".
Internal helper contract for calling the multi-cli-companion runtime from any multi:* subagent
Internal guidance for presenting multi:* helper output back to the user
Auto-detect plan files in conversation context and pass them to multi:* execute subagents by reference, not by paraphrase
Internal guidance for presenting Codex helper output back to the user
| name | codex-cli-runtime |
| description | Internal helper contract for calling the multi-cli-companion runtime from Claude Code |
| user-invocable | false |
Use this skill only inside the multi:codex-rescue subagent.
Primary helper:
node "${CLAUDE_PLUGIN_ROOT}/scripts/multi-cli-companion.mjs" task --cli codex "<raw arguments>"Execution rules:
task once and return that stdout unchanged.git, direct Codex CLI strings, or any other Bash activity.setup, review, adversarial-review, status, result, or cancel from multi:codex-rescue.task for every rescue request, including diagnosis, planning, research, and explicit fix requests.gpt-5-4-prompting skill to rewrite the user's request into a tighter Codex prompt before the single task call.--effort unset unless the user explicitly requests a specific effort.--model only when the user explicitly asks for one.spark to --model gpt-5.3-codex-spark.--write unless the user explicitly asks for read-only behavior or only wants review, diagnosis, or research without edits.Command selection:
task invocation per rescue handoff.task in the FOREGROUND (no --background) so the call returns Codex's real result; the parent command handles background scheduling by running this subagent as a harness background task (that is what notifies the main thread). --background / --wait are Claude-side scheduling controls — strip them from the natural-language task text; only honor an explicit --background as user-requested fire-and-forget polled via /multi:status.--model, normalize spark to gpt-5.3-codex-spark and pass it through to task.--effort, pass it through to task.--resume, strip that token from the task text and add --resume-last.--fresh, strip that token from the task text and do not add --resume-last.--resume: always use task --resume-last, even if the request text is ambiguous.--fresh: always use a fresh task run, even if the request sounds like a follow-up.--effort: accepted values are none, minimal, low, medium, high, xhigh.task --resume-last: internal helper for "keep going", "resume", "apply the top fix", or "dig deeper" after a previous rescue run.Safety rules:
multi:codex-rescue unless the user explicitly asks for read-only behavior.task command exactly as-is.