docs(knowledge): compare routines-equivalent cloud agents across the 4 CLIs#231
Conversation
…4 CLIs Add a cross-ecosystem comparison (Claude Code Routines / OpenAI Codex cloud + Workspace agents / Google Jules / GitHub Copilot coding agent) to scheduled-tasks.md, scored on cloud-async, schedule/API/GitHub triggers, and autonomous PR. Verified trigger types against each vendor's official docs; uncertain cells marked as such. Track the new Jules and Copilot coding-agent docs in staleness sources. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
…upport Confirmed Codex cloud, Jules (scheduled-tasks + API reference) and the Copilot coding agent against each vendor's official docs. Jules actually bundles all three triggers (jules label, v1alpha REST API, scheduled tasks) — correcting the earlier "unknown" cells. Note Workspace agents stays KB-sourced because OpenAI's pages block automated fetch (403). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
🔍 drive review (mode: quick)適用観点 (6/11):
correctness指摘なし。比較表の各セルは公式 docs と一致(Codex cloud=web/IDE/ security指摘なし。 conventions指摘なし。 compatibility指摘なし。 usability指摘なし。凡例と「ソース確認状況」注記で表の読み方・確度が明快。 documentation指摘なし。403 で取得不能だった Workspace agents を隠さず明記し、 サマリーCritical: 0 件 / Warning: 0 件 / Info: 0 件 判定: 全適用観点が exit_criteria を満たす → merge-ready |
概要
「Claude 以外に Routines 類似機能はあるか」を調査し、主要 4 コーディングエージェント CLI(Claude Code / Codex CLI / Gemini CLI / GitHub Copilot CLI)を中心に、各エコシステムの「Routines 相当(クラウド自律/定期実行)」比較表を
ai/practice/scheduled-tasks.mdの section 5 に追加。調査結果(要点)
julesラベル→PR、experimental)。スケジュール/公開 API は公式 docs 未確認。検証
不明/△と明記。sources.yamlに追加。🤖 Generated with Claude Code