{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreia63pxektqon243rx5jjtywaon5kl7uyjb6467zok4ru56rmkca3u",
    "uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3mlsk6dktwnp2"
  },
  "path": "/t/proposal-hybrid-codex-cli-mode-that-combines-coding-with-interactive-planning/1380506#post_2",
  "publishedAt": "2026-05-14T09:20:40.000Z",
  "site": "https://community.openai.com",
  "textContent": "I think this idea could also be useful as an optional setting for `/goal` mode.\n\nI understand that `/goal` is designed for continuous autonomous work until the target result is achieved. That is a strong and useful concept. However, full autonomy is not always the best choice.\n\nSometimes a task reaches a point where a human decision is needed. Without that checkpoint, the agent may continue trying potentially wrong approaches again and again, either failing to reach the result or eventually producing a result based on an incorrect assumption.\n\nSo it would be useful to have two ways to run `/goal`:\n\n  * full autonomous `/goal` mode, when the user fully trusts the agent to make decisions and keep working until the result is reached\n\n  * `/goal` with critical-decision confirmations, when the user wants the agent to continue working autonomously but pause before important choices\n\n\n\n\nFor example, the agent could ask for confirmation before changing architecture, adding dependencies, modifying database schema, applying destructive changes, or choosing between several unclear implementation paths.\n\nThis would preserve the main purpose of `/goal`, but give users more control when the task contains important ambiguity.\n\nIn other words:\n\nIf I fully trust the agent, I run `/goal` in full autonomous mode.\n\nIf I want control over key decisions, I run `/goal` with confirmation checkpoints.",
  "title": "Proposal: Hybrid Codex CLI mode that combines coding with interactive planning"
}