メニュー
Monitor Codex CLI session status, detect timeouts, errors, and completion. Use when user asks about Codex session status, wants to check if Codex is still running, needs to debug a stalled session, or mentions "codex monitor", "session status", "codex 状态", "session 断了吗".
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | codex-monitor |
| description | Monitor Codex CLI session status, detect timeouts, errors, and completion. Use when user asks about Codex session status, wants to check if Codex is still running, needs to debug a stalled session, or mentions "codex monitor", "session status", "codex 状态", "session 断了吗". |
Monitor and analyze Codex CLI sessions from ~/.codex/sessions/.
Run the monitor script:
python3 <skill-dir>/scripts/codex-monitor.py [today|yesterday|YYYY-MM-DD] [--watch] [--last N]
| Command | What it does |
|---|---|
python3 codex-monitor.py | Show all sessions from today |
python3 codex-monitor.py --watch | Live monitor (refresh every 10s, Ctrl+C to stop) |
python3 codex-monitor.py --last 5 | Show only the last 5 sessions |
python3 codex-monitor.py yesterday | Show yesterday's sessions |
python3 codex-monitor.py 2026-02-20 | Show a specific date |
| Icon | Meaning | Action |
|---|---|---|
| ✅ | Completed normally | None needed |
| 🔄 | Still running | Wait or check progress |
| ⏱️ | Timeout/stale (>10min no output) | Likely dead, restart task |
| ⚠️ | Possibly stale (2-10min) | Check soon |
| ❌ | Error | Read error message |
| 📭 | Empty/initialized only | Session didn't start |
Look for 🔄 (running) status. If you see ⏱️ (timeout), the session likely died.
Look for ✅ (completed) and check last_message for the final output summary.
Check for:
The script shows parent-child relationships with indentation. A main session spawns sub-agents (子Agent) for parallel work. Check if all children completed.
Manage long-running, multi-hour AI agent tasks using durable project memory — a structured 4-file system (SPEC, PLAN, RULES, STATUS) that prevents drift, ensures verification, and keeps work coherent across extended sessions. Use when the user asks to build something complex from scratch, run a multi-milestone project, delegate large autonomous coding work, set up a long-running agent session, create a project with 5+ milestones, or mentions "long horizon", "长任务", "长时间运行", "多里程碑", "自主编码", "autonomous coding", "durable memory". Based on OpenAI's Codex 25-hour design-tool experiment documented at https://developers.openai.com/cookbook/examples/codex/long_horizon_tasks.
双 Agent 全面审查修复框架。Antigravity + Codex MCP 独立审查 → 交叉验证 → 协作修复。内置 Codex session 监控、权限管控、双向状态追踪、清理策略。触发词："Ralph Loop"、"raphaloop"、"全面审查"、"零容错审查"、"/ralph-loop"。适用场景：代码审查、数据质量检查、发布前终验、跨 Agent 协作修复。