codex-lark
muggle-stack/codex-lark
Codex-lark is a local-first bridge connecting Lark (Feishu) messaging directly to Codex and Claude Code engines, enabling chat-based task execution and secure knowledge agent routing.
3 2MITUpdated 2026-07-24
74
GoodOverview
This skill/integration provides a bridge between Lark/Feishu and local AI development engines (Codex and Claude Code). It listens to authorized user messages via whitelists, dispatches tasks to AI agents, renders native dynamic progress cards in Lark chats, and manages session mapping and read-only knowledge agents for colleagues.
Capabilities
- ▸Lark message event integration and parsing
- ▸Pluggable engine support for Codex CLI and Claude Code
- ▸Native dynamic progress card rendering in Lark
- ▸Per-user session management and aliasing
- ▸Role-based permission mapping (read-only, workspace-write, danger-full-access)
- ▸OS-level sandboxing (bubblewrap) for read-only knowledge agents
- ▸File and image attachment processing from user messages
Best for
Triggering Codex tasks directly from Lark direct messages or group chats by mentioning the bot, Operating as an admin/owner to handle cross-repository coding, testing, and documentation tasks remotely, Deploying a secure, read-only knowledge agent for teammates to query internal documentation and skills, Managing long-running AI sessions and aliases directly through Lark chat commands
Works with
Claude CodeCodex
Evoa Score breakdown= Σ (score × weight)
74Task usefulness20%100 → +20.0
実タスクにどれだけ役立つか（機能の豊富さ・用途の明確さ）。 — AIによるcapabilities/use-cases解析
Code quality15%85 → +12.8
実装・指示の品質。 — AIによるSKILL.md/README解析
Maintenance15%100 → +15.0
リポジトリがどれだけ活発に保守されているか。 — GitHub 最終push日時の新しさ
Documentation12%41 → +4.9
ドキュメントの充実度・分かりやすさ。 — README/独自要約の情報量
Security15%50 → +7.5
危険・不審な挙動が無いか。 — AIによるセキュリティレビュー
Originality10%90 → +9.0
ありふれたラッパーではない独自性。 — AIによる独自性判定
Popularity8%23 → +1.8
コミュニティの採用度。 — GitHub Stars/Forks（対数スケール）
Compatibility5%50 → +2.5
対応AIエージェントの広さ。 — AIによる対応エージェント判定
Weighted total73.5 / 100
ライセンス不明/制限あり（Red）のSkillは総合スコアに0.85倍の補正を適用します。 ランキングはこのScoreのみで決まり、広告で変わりません。 算出方法の詳細 →
Security considerations
Requires strict management of owner IDs and colleague allowlists with LARK_CODEX_ALLOW_ALL set to 0. Colleague agents use read-only sandboxing via bubblewrap to prevent unauthorized modifications or credential leaks.
Categories
Summary and analysis are original content generated by AI Skills Rank. The skill's source text is not reproduced here — view it on the linked repository.