oh-my-opencode 4.7.5 → 4.8.1

package/README.ja.md

@@ -1,11 +1,11 @@
 > [!NOTE]
-> **🚀 初の Codex リリース：omo が Codex CLI で利用可能になりました**
+> **OmO for Codex が登場しました。LazyCodex を試してください**
 >
-> 複雑な JSON 設定はもう不要です。以下を実行するだけ：
+> Anthropic のモデルを愛しすぎてブロックまで経験した私たちは、今度は Codex に賭けることにしました。
+> OmO のファンだけれど設定が面倒だったなら、LazyCodex を使ってください。OmO for Codex はもうリリースされています:
 > ```bash
 > npx lazycodex-ai install
 > ```
-> あなたの Codex は明確にコーディングし、組み込みの computer-use QA で作業を完了します。
 > 詳細は [lazycodex.ai](https://lazycodex.ai) をご覧ください。
 
 > [!NOTE]
@@ -133,7 +133,7 @@ curl -s https://raw.githubusercontent.com/code-yeongyu/oh-my-openagent/refs/head
 
 匿名のテレメトリは、アクティブなインストール数(DAU/WAU/MAU)の集計のためにデフォルトで有効になっています。マシン1台につきUTC日あたり最大1回イベントが送信され、ハッシュ化されたインストール識別子を使用し、生のホスト名は使用せず、PostHog person profile も作成されません。無効化するには `OMO_SEND_ANONYMOUS_TELEMETRY=0` または `OMO_DISABLE_POSTHOG=1` を設定してください。[プライバシーポリシー](docs/legal/privacy-policy.md)と[利用規約](docs/legal/terms-of-service.md)をご覧ください。
 
-**Ultimate と Light:** oh-my-openagent は同じ製品の 2 つのエディションとして提供されます。**Ultimate エディション**（`bunx oh-my-openagent install` または `--platform=opencode`、デフォルト）は OpenCode 上のフル機能で、11 エージェント、54+ フック、Team Mode、すべての MCP、スラッシュコマンド、IntentGate モードを提供します。**Light エディション**（`bunx oh-my-openagent install --platform=codex`）は OpenAI Codex CLI のプラグインシステムへ綺麗に移植できる 5 コンポーネント（`rules`、`comment-checker`、`lsp`、`ultrawork`、`ulw-loop`）のみを提供します。`bunx lazycodex-ai install` は `--platform=codex` のショートカット別名です。両方を同時にインストールするには `--platform=both`。Codex 専用テレメトリは `OMO_CODEX_DISABLE_POSTHOG=1` または `OMO_CODEX_SEND_ANONYMOUS_TELEMETRY=0` で無効化できます。
+**Ultimate と Light:** oh-my-openagent は同じ製品の 2 つのエディションとして提供されます。**Ultimate エディション**（`bunx oh-my-openagent install` または `--platform=opencode`、デフォルト）は OpenCode 上のフル機能で、11 エージェント、54+ フック、Team Mode、すべての MCP、スラッシュコマンド、IntentGate モードを提供します。**Light エディション**（`npx lazycodex-ai install` または `bunx oh-my-openagent install --platform=codex`）は OpenAI Codex CLI のプラグインシステムへ綺麗に移植できる 8 コンポーネント（`rules`、`comment-checker`、`git-bash`、`lsp`、`ultrawork`、`ulw-loop`、`start-work-continuation`、`telemetry`）を提供します。両方を同時にインストールするには `--platform=both`。Codex 専用テレメトリは `OMO_CODEX_DISABLE_POSTHOG=1` または `OMO_CODEX_SEND_ANONYMOUS_TELEMETRY=0` で無効化できます。
 
 ---
 
@@ -165,14 +165,14 @@ Read this and tell me why it's not just another boilerplate: https://raw.githubu
 |       | 機能                                                     | Editions | 何をするのか                                                                                                                                                                                                                   |
 | :---: | :------------------------------------------------------- | :------: | :----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
 |   🤖   | **規律あるエージェント (Discipline Agents)**             | Ultimate | Sisyphus が Hephaestus、Oracle、Librarian、Explore をオーケストレーションします。完全な AI 開発チームが並列で動きます。                                                                                                        |
-|   🧩   | **Codex CLI Light Edition**                              | Light    | OpenAI Codex CLI 上で動作する omo の 5 つの移植コンポーネント (rules, comment-checker, LSP, ultrawork, ulw-loop)。インストール: `bunx oh-my-openagent install --platform=codex`。                                              |
+|   🧩   | **Codex CLI Light Edition**                              | Light    | OpenAI Codex CLI 上で動作する omo の 8 つの移植コンポーネント (rules, comment-checker, git-bash, LSP, ultrawork, ulw-loop, start-work continuation, telemetry)。インストール: `npx lazycodex-ai install`。                                    |
 |   👥   | **Team Mode** (v4.0, オプトイン)                          | Ultimate | リードエージェント + 最大 8 メンバーの並列実行、リアルタイム tmux 可視化、専用 `team_*` ツール群。`hyperplan`(5 人の敵対的批評家)と `security-research`(3 人のハンター + 2 人の PoC エンジニア)を駆動します。[ドキュメント →](docs/guide/team-mode.md) |
 |   ⚡   | **`ultrawork` / `ulw`**                                  | Both     | 一言で OK。すべてのエージェント (Ultimate) または Codex `ultrawork` コンポーネント (Light) がアクティブになり、終わるまで止まりません。                                                                                          |
 |   🚪   | **[IntentGate](https://factory.ai/news/terminal-bench)** | Ultimate | ユーザーの真の意図を分析してから分類・行動します。`search` / `analyze` / `team` / `hyperplan` をトリガー。(Light は `ulw` / `ultrawork` のみフック。)                                                                          |
 |   🔗   | **ハッシュベースの編集ツール**                           | Ultimate | `LINE#ID` のコンテンツハッシュですべての変更を検証します。stale-line エラー 0%。[oh-my-pi](https://github.com/can1357/oh-my-pi) にインスパイア。[The Harness Problem →](https://blog.can.ac/2026/02/12/the-harness-problem/) (Codex はネイティブの `apply_patch` を使用。) |
-|   🛠️   | **LSP + AST-Grep**                                       | Ultimate | ワークスペース単位のリネーム、ビルド前の診断、AST を考慮した書き換え。エージェントに IDE レベルの精度を提供。(LSP は Light でも `lsp` コンポーネントで動作; AST-Grep は Ultimate のみ。)                                       |
+|   🛠️   | **LSP + AST-Grep**                                       | Both     | ワークスペース単位のリネーム、ビルド前の診断、AST を考慮した書き換え。Ultimate は組み込みツールとして、Light は plugin-scoped `lsp` / `ast_grep` MCP として提供します。                                       |
 |   🧠   | **バックグラウンドエージェント**                         | Ultimate | 5 人以上の専門家を並列で投入。コンテキストは軽く保ち、結果は準備ができ次第受け取ります。                                                                                                                                       |
-|   📚   | **組み込み MCP**                                         | Ultimate | Exa (Web 検索)、Context7 (公式ドキュメント)、Grep.app (GitHub 検索)。常にオン。(Light は LSP MCP のみ。)                                                                                                                       |
+|   📚   | **組み込み MCP**                                         | Both     | Ultimate は Exa (Web 検索)、Context7 (公式ドキュメント)、Grep.app (GitHub 検索) をランタイム注入します。Light は plugin-scoped MCP として `ast_grep`、`grep_app`、`context7`、`git_bash`、`lsp` を提供します。                                                                                                                       |
 |   🔁   | **Ralph Loop / `/ulw-loop`**                             | Ultimate | 自己参照ループ。100% 完了するまで絶対に止まりません。                                                                                                                                                                          |
 |   ✅   | **Todo Enforcer** (Boulder)                              | Ultimate | エージェントがサボる？システムが首根っこを掴んで戻します。あなたのタスクは必ず終わります。                                                                                                                                     |
 |   💬   | **コメントチェッカー**                                   | Both     | コメントから AI 臭い無駄話を排除。両エディションで同じ `@code-yeongyu/comment-checker` バイナリが動作。                                                                                                                       |

package/README.ko.md

@@ -1,11 +1,11 @@
 > [!NOTE]
-> **🚀 첫 Codex 버전 출시: omo 이제 Codex CLI에서 사용 가능합니다**
+> **OmO for Codex 출시: LazyCodex를 써보세요**
 >
-> 더 이상 복잡한 JSON 설정은 필요 없습니다. 그냥 실행하세요:
+> Anthropic 모델을 너무 사랑했던 이유로 차단까지 겪었던 저희는 이제 Codex의 손을 들기로 했습니다.
+> OmO 팬이지만 설정이 번거로웠다면 LazyCodex를 사용해보세요. OmO for Codex가 출시되었습니다:
 > ```bash
 > npx lazycodex-ai install
 > ```
-> 당신의 Codex가 명확하게 코딩하고, 내장 computer-use QA로 작업을 완수합니다.
 > 자세한 내용은 [lazycodex.ai](https://lazycodex.ai)에서 확인하세요.
 
 > [!NOTE]
@@ -134,7 +134,7 @@ curl -s https://raw.githubusercontent.com/code-yeongyu/oh-my-openagent/refs/head
 
 익명 텔레메트리는 활성 설치 수(DAU/WAU/MAU) 집계를 위해 기본적으로 활성화되어 있습니다. 머신당 UTC 하루에 최대 1회만 이벤트가 전송되며, 해시된 설치 식별자를 사용하고 원시 호스트명은 절대 사용하지 않으며 PostHog person profile은 생성되지 않습니다. `OMO_SEND_ANONYMOUS_TELEMETRY=0` 또는 `OMO_DISABLE_POSTHOG=1`로 비활성화할 수 있습니다. [개인정보처리방침](docs/legal/privacy-policy.md)과 [서비스 이용약관](docs/legal/terms-of-service.md)을 참조하세요.
 
-**Ultimate vs Light:** oh-my-openagent는 같은 제품의 두 에디션으로 출시됩니다. **Ultimate 에디션**(`bunx oh-my-openagent install` 또는 `--platform=opencode`, 기본값)은 OpenCode 위에서 풀 기능 — 11 agent, 54+ hook, Team Mode, 모든 MCP, 슬래시 명령, IntentGate 모드 — 을 제공합니다. **Light 에디션**(`bunx oh-my-openagent install --platform=codex`)은 OpenAI Codex CLI의 플러그인 시스템에 깔끔히 포팅되는 5개 컴포넌트(`rules`, `comment-checker`, `lsp`, `ultrawork`, `ulw-loop`)만 제공합니다. `bunx lazycodex-ai install`은 `--platform=codex`의 단축 별칭입니다. 둘 다 설치하려면 `--platform=both`. Codex 전용 텔레메트리는 `OMO_CODEX_DISABLE_POSTHOG=1` 또는 `OMO_CODEX_SEND_ANONYMOUS_TELEMETRY=0`으로 비활성화할 수 있습니다.
+**Ultimate vs Light:** oh-my-openagent는 같은 제품의 두 에디션으로 출시됩니다. **Ultimate 에디션**(`bunx oh-my-openagent install` 또는 `--platform=opencode`, 기본값)은 OpenCode 위에서 풀 기능 — 11 agent, 54+ hook, Team Mode, 모든 MCP, 슬래시 명령, IntentGate 모드 — 을 제공합니다. **Light 에디션**(`npx lazycodex-ai install` 또는 `bunx oh-my-openagent install --platform=codex`)은 OpenAI Codex CLI의 플러그인 시스템에 깔끔히 포팅되는 8개 컴포넌트(`rules`, `comment-checker`, `git-bash`, `lsp`, `ultrawork`, `ulw-loop`, `start-work-continuation`, `telemetry`)를 제공합니다. 둘 다 설치하려면 `--platform=both`. Codex 전용 텔레메트리는 `OMO_CODEX_DISABLE_POSTHOG=1` 또는 `OMO_CODEX_SEND_ANONYMOUS_TELEMETRY=0`으로 비활성화할 수 있습니다.
 
 ---
 
@@ -166,14 +166,14 @@ Read this and tell me why it's not just another boilerplate: https://raw.githubu
 |       | 기능                                                      | Editions | 하는 일                                                                                                                                                                                                          |
 | :---: | :------------------------------------------------------- | :------: | :--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
 |   🤖   | **Discipline Agents**                                    | Ultimate | Sisyphus가 Hephaestus, Oracle, Librarian, Explore를 지휘합니다. 병렬로 도는 풀스택 AI 개발팀.                                                                                                                    |
-|   🧩   | **Codex CLI Light Edition**                              | Light    | OpenAI Codex CLI에서 동작하는 omo의 5개 포팅 컴포넌트(rules, comment-checker, LSP, ultrawork, ulw-loop). 설치: `bunx oh-my-openagent install --platform=codex`.                                                              |
+|   🧩   | **Codex CLI Light Edition**                              | Light    | OpenAI Codex CLI에서 동작하는 omo의 8개 포팅 컴포넌트(rules, comment-checker, git-bash, LSP, ultrawork, ulw-loop, start-work continuation, telemetry). 설치: `npx lazycodex-ai install`.                                                   |
 |   👥   | **Team Mode** (v4.0, opt-in)                             | Ultimate | 리드 에이전트 + 최대 8명의 병렬 멤버, 실시간 tmux 시각화, 전용 `team_*` 도구. `hyperplan`(5명의 적대적 비평가)과 `security-research`(3명의 헌터 + 2명의 PoC 엔지니어)를 구동합니다. [문서 →](docs/guide/team-mode.md) |
 |   ⚡   | **`ultrawork` / `ulw`**                                  | Both     | 한 단어. 모든 에이전트(Ultimate)나 Codex `ultrawork` 컴포넌트(Light)가 켜집니다. 끝날 때까지 멈추지 않습니다.                                                                                                    |
 |   🚪   | **[IntentGate](https://factory.ai/news/terminal-bench)** | Ultimate | 분류하거나 행동하기 전에 사용자의 진짜 의도부터 분석합니다. `search` / `analyze` / `team` / `hyperplan` 트리거. (Light는 `ulw` / `ultrawork`만 hook.)                                                            |
 |   🔗   | **Hash-Anchored Edit Tool**                              | Ultimate | `LINE#ID` 콘텐츠 해시가 모든 변경을 검증합니다. 낡은 라인 에러 0건. [oh-my-pi](https://github.com/can1357/oh-my-pi)에서 영감. [The Harness Problem →](https://blog.can.ac/2026/02/12/the-harness-problem/) (Codex는 자체 `apply_patch` 사용.) |
-|   🛠️   | **LSP + AST-Grep**                                       | Ultimate | 워크스페이스 리네임, 빌드 전 진단, AST 기반 리라이트. 에이전트에게도 IDE 수준의 정밀도. (LSP는 Light에서도 `lsp` 컴포넌트로 동작; AST-Grep은 Ultimate 전용.)                                                     |
+|   🛠️   | **LSP + AST-Grep**                                       | Both     | 워크스페이스 리네임, 빌드 전 진단, AST 기반 리라이트. Ultimate는 내장 도구로, Light는 plugin-scoped `lsp` / `ast_grep` MCP로 제공합니다.                                                     |
 |   🧠   | **Background Agents**                                    | Ultimate | 전문가 5명 이상을 동시에 발사. 컨텍스트는 가볍게. 결과는 준비되면 도착.                                                                                                                                          |
-|   📚   | **Built-in MCPs**                                        | Ultimate | Exa(웹 검색), Context7(공식 문서), Grep.app(GitHub 검색). 항상 켜져 있음. (Light는 LSP MCP만.)                                                                                                                   |
+|   📚   | **Built-in MCPs**                                        | Both     | Ultimate는 Exa(웹 검색), Context7(공식 문서), Grep.app(GitHub 검색)를 런타임에 주입합니다. Light는 plugin-scoped MCP로 `ast_grep`, `grep_app`, `context7`, `git_bash`, `lsp`를 제공합니다.                                                                                                                   |
 |   🔁   | **Ralph Loop / `/ulw-loop`**                             | Ultimate | 자기참조 루프. 100% 끝날 때까지 멈추지 않습니다.                                                                                                                                                                 |
 |   ✅   | **Todo Enforcer** (Boulder)                              | Ultimate | 에이전트가 놀고 있나요? 시스템이 다시 끌어옵니다. 당신의 작업은 반드시 끝납니다.                                                                                                                                |
 |   💬   | **Comment Checker**                                      | Both     | 주석에 AI 슬롭 금지. 동일한 `@code-yeongyu/comment-checker` 바이너리가 두 에디션 모두에서 동작.                                                                                                                  |
@@ -441,6 +441,8 @@ OpenCode가 Debian/Arch라면, oh-my-openagent는 Ubuntu/[Omarchy](https://omarc
 
 언급된 어떤 프로젝트나 모델과도 제휴 관계는 없습니다. 그저 개인적인 실험의 결과입니다.
 
+Credit: LazyCodex 이름 아이디어는 [LazyVim](https://github.com/LazyVim/LazyVim)에서 영감을 받았습니다. Ultragoal과 UltraQA 아이디어는 [oh-my-codex](https://github.com/Yeachan-Heo/oh-my-codex)에서 영감을 받아 OmO용으로 콘셉트만 재구현했습니다.
+
 이 프로젝트의 99%는 OpenCode로 만들어졌습니다. 저는 TypeScript를 사실 잘 모릅니다. **다만 이 문서만큼은 제가 직접 검토하고 대부분 다시 썼습니다.**
 
 ## 전문가들이 현업에서 쓰고 있습니다

package/README.md

@@ -1,11 +1,11 @@
 > [!NOTE]
-> **🚀 First Codex Release: omo is now available for Codex CLI**
+> **OmO for Codex is here: try LazyCodex**
 >
-> No more complex JSON configurations. Just run:
+> We loved Anthropic models enough to get blocked. Now we are backing Codex.
+> If you are an OmO fan but the setup felt like too much, use LazyCodex. OmO for Codex has shipped:
 > ```bash
 > npx lazycodex-ai install
 > ```
-> Your Codex will code clearly and complete tasks using built-in computer-use QA.
 > Learn more at [lazycodex.ai](https://lazycodex.ai).
 
 > [!NOTE]
@@ -114,7 +114,7 @@ Install oh-my-openagent. Type `ultrawork`. Done.
 oh-my-openagent ships in two editions of the same product:
 
 - **Ultimate Edition (omo for OpenCode)** — full omo. 11 agents, 54+ lifecycle hooks, 5 built-in MCPs, all slash commands, Team Mode, ulw-loop, ultrawork, hashline edits — everything.
-- **Light Edition (omo for Codex CLI)** — the portable components that fit Codex's plugin system: `rules`, `comment-checker`, `lsp`, `ultrawork`, `ulw-loop`, `start-work-continuation`, and `telemetry`. No agent orchestration, no `team_*` tools, no built-in MCPs beyond LSP — Codex CLI's own surface does that work.
+- **Light Edition (omo for Codex CLI)** — the portable components that fit Codex's plugin system: `rules`, `comment-checker`, `git-bash`, `lsp`, `ultrawork`, `ulw-loop`, `start-work-continuation`, and `telemetry`, plus plugin-scoped MCPs for `ast_grep`, `grep_app`, `context7`, `git_bash`, and `lsp`. No agent orchestration and no `team_*` tools — Codex CLI's own surface does that work.
 
 Pick the edition(s) you want.
 
@@ -157,7 +157,7 @@ Fetch the full guide and follow it step by step:
 curl -fsSL https://raw.githubusercontent.com/code-yeongyu/oh-my-openagent/refs/heads/dev/docs/guide/installation.md
 ```
 
-The guide covers: platform selection, the subscription interview, provider authentication (Anthropic / Gemini / Copilot / Z.ai / OpenCode Zen), the agent-to-model matching matrix, modes (`ultrawork`, `search`, `analyze`, `team`, `hyperplan`), slash commands, the Light edition's 5 Codex components, Team Mode, and uninstall. Don't summarize it; read it end to end.
+The guide covers: platform selection, the subscription interview, provider authentication (Anthropic / Gemini / Copilot / Z.ai / OpenCode Zen), the agent-to-model matching matrix, modes (`ultrawork`, `search`, `analyze`, `team`, `hyperplan`), slash commands, the Light edition's 8 Codex components, Team Mode, and uninstall. Don't summarize it; read it end to end.
 
 ### Note on package and command names
 
@@ -212,7 +212,7 @@ Even with only the following subscriptions, `ultrawork` works well (this project
 |       | Feature                                                  | Edition  | What it does                                                                                                                                                                                                     |
 | :---: | :------------------------------------------------------- | :------: | :--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
 |   🤖   | **Discipline Agents**                                    | Ultimate | Sisyphus orchestrates Hephaestus, Oracle, Librarian, Explore. A full AI dev team in parallel.                                                                                                                    |
-|   🧩   | **Codex CLI Light Edition**                              | Light    | Portable OMO components (rules, comment-checker, LSP, ultrawork, ulw-loop, start-work continuation, telemetry) running inside OpenAI Codex CLI. Install via `npx lazycodex-ai install`.                       |
+|   🧩   | **Codex CLI Light Edition**                              | Light    | Portable OMO components (rules, comment-checker, git-bash, LSP, ultrawork, ulw-loop, start-work continuation, telemetry) running inside OpenAI Codex CLI. Install via `npx lazycodex-ai install`.             |
 |   👥   | **Team Mode** (v4.0, opt-in)                             | Ultimate | Lead agent + up to 8 parallel members, real-time tmux visualization, dedicated `team_*` tools. Powers `hyperplan` (5 hostile critics) and `security-research` (3 hunters + 2 PoC engineers). [Docs →](docs/guide/team-mode.md) |
 |   ⚡   | **`ultrawork` / `ulw`**                                  | Both     | One word. Every agent activates. Doesn't stop until done.                                                                                                                                                        |
 |   🚪   | **[IntentGate](https://factory.ai/news/terminal-bench)** | Ultimate | Analyzes true user intent before classifying or acting. No more literal misinterpretations. (Light edition only recognises the `ultrawork`/`ulw` keyword.)                                                       |
@@ -407,12 +407,16 @@ To remove oh-my-openagent:
 4. **Remove omo-codex (Codex CLI Light edition)**
 
    ```bash
+   npx lazycodex-ai uninstall
+   # backward-compatible alias:
    npx lazycodex-ai cleanup
-   # or, from the shared CLI:
+
+   omo uninstall --platform=codex
+   # backward-compatible alias:
    omo cleanup --platform=codex
    ```
 
-   The cleanup command removes managed `sisyphuslabs` Codex cache/marketplace state, strips `omo@sisyphuslabs` plugin and hook-state blocks from `~/.codex/config.toml` after writing a backup, and removes agent TOML links listed in the install manifest. If a specific project still has old `oh-my-codex` / `omx` state, run the command from that project or pass `--project <path>`; it repairs the known project-local `.codex/config.toml` conflict and reports `.codex` / `.omx` artifacts without deleting project-owned files.
+   The uninstall command removes managed `sisyphuslabs` Codex cache/marketplace state, strips `omo@sisyphuslabs` plugin and hook-state blocks from `~/.codex/config.toml` after writing a backup, and removes agent TOML links listed in the install manifest. If a specific project still has old project-local Codex plugin state, run the command from that project or pass `--project <path>`; it repairs known project-local `.codex/config.toml` conflicts and reports project-local `.codex` artifacts without deleting project-owned files.
 
 ## Features
 
@@ -490,6 +494,8 @@ Sounds arrogant? Have a better way? Contribute. You're welcome.
 
 No affiliation with any project or model mentioned. Just personal experimentation.
 
+Credit: The LazyCodex name idea is inspired by [LazyVim](https://github.com/LazyVim/LazyVim). The Ultragoal, UltraQA, and Sparkshell ideas are inspired by [oh-my-codex](https://github.com/Yeachan-Heo/oh-my-codex), reimplemented from concept for OmO.
+
 99% of this project was built with OpenCode. I don't really know TypeScript, **but I personally reviewed and largely rewrote this doc.**
 
 ## Loved by professionals at

package/README.ru.md

@@ -5,7 +5,7 @@
 > ```bash
 > npx lazycodex-ai install
 > ```
-> Ваш Codex будет писать чистый код и завершать задачи с помощью встроенного computer-use QA.
+> Ваш Codex получит локальные правила, comment checker, LSP, Git Bash для Windows, ultrawork и ulw-loop.
 > Подробности на [lazycodex.ai](https://lazycodex.ai).
 
 > [!NOTE]
@@ -131,7 +131,7 @@ curl -s https://raw.githubusercontent.com/code-yeongyu/oh-my-openagent/refs/head
 
 Анонимная телеметрия включена по умолчанию для подсчёта активных установок (DAU/WAU/MAU). Не более одного события на машину за UTC-сутки, использует хешированный идентификатор установки, никогда не использует исходное имя хоста, и не создаёт PostHog person profile. Можно отключить через `OMO_SEND_ANONYMOUS_TELEMETRY=0` или `OMO_DISABLE_POSTHOG=1`. См. [Политику конфиденциальности](docs/legal/privacy-policy.md) и [Условия обслуживания](docs/legal/terms-of-service.md).
 
-**Ultimate и Light:** oh-my-openagent поставляется в двух редакциях одного продукта. **Ultimate** (`bunx oh-my-openagent install` или `--platform=opencode`, по умолчанию) — полнофункциональная редакция поверх OpenCode: 11 агентов, 54+ хука, Team Mode, все MCP, все слэш-команды, режимы IntentGate. **Light** (`bunx oh-my-openagent install --platform=codex`) — только 5 компонентов omo, которые портируются в систему плагинов OpenAI Codex CLI: `rules`, `comment-checker`, `lsp`, `ultrawork`, `ulw-loop`. `bunx lazycodex-ai install` — это сокращённый псевдоним для `--platform=codex`. Чтобы установить обе редакции одной командой, используйте `--platform=both`. Телеметрию только для Codex можно отключить через `OMO_CODEX_DISABLE_POSTHOG=1` или `OMO_CODEX_SEND_ANONYMOUS_TELEMETRY=0`.
+**Ultimate и Light:** oh-my-openagent поставляется в двух редакциях одного продукта. **Ultimate** (`bunx oh-my-openagent install` или `--platform=opencode`, по умолчанию) — полнофункциональная редакция поверх OpenCode: 11 агентов, 54+ хука, Team Mode, все MCP, все слэш-команды, режимы IntentGate. **Light** (`npx lazycodex-ai install` или `bunx oh-my-openagent install --platform=codex`) — 8 компонентов omo, которые портируются в систему плагинов OpenAI Codex CLI: `rules`, `comment-checker`, `git-bash`, `lsp`, `ultrawork`, `ulw-loop`, `start-work-continuation`, `telemetry`. Чтобы установить обе редакции одной командой, используйте `--platform=both`. Телеметрию только для Codex можно отключить через `OMO_CODEX_DISABLE_POSTHOG=1` или `OMO_CODEX_SEND_ANONYMOUS_TELEMETRY=0`.
 
 ------
 
@@ -164,14 +164,14 @@ Read this and tell me why it's not just another boilerplate: https://raw.githubu
 |     | Функция                                                  | Editions | Что делает                                                                                                                                                                                                                       |
 | --- | -------------------------------------------------------- | :------: | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
 | 🤖   | **Дисциплинированные агенты**                            | Ultimate | Sisyphus оркестрирует Hephaestus, Oracle, Librarian, Explore. Полноценная AI-команда разработки в параллельном режиме.                                                                                                           |
-| 🧩   | **Codex CLI Light Edition**                              | Light    | 5 компонентов omo, портированных в OpenAI Codex CLI (rules, comment-checker, LSP, ultrawork, ulw-loop). Установка: `bunx oh-my-openagent install --platform=codex`.                                                                          |
+| 🧩   | **Codex CLI Light Edition**                              | Light    | 8 компонентов omo, портированных в OpenAI Codex CLI (rules, comment-checker, git-bash, LSP, ultrawork, ulw-loop, start-work continuation, telemetry). Установка: `npx lazycodex-ai install`.                                                                |
 | 👥   | **Team Mode** (v4.0, opt-in)                             | Ultimate | Лид-агент + до 8 параллельных участников, визуализация в tmux в реальном времени, выделенные инструменты `team_*`. Питает `hyperplan` (5 враждебных критиков) и `security-research` (3 охотника + 2 PoC-инженера). [Документация →](docs/guide/team-mode.md) |
 | ⚡   | **`ultrawork` / `ulw`**                                  | Both     | Одно слово. Все агенты (Ultimate) или Codex-компонент `ultrawork` (Light) активируются. Не останавливается, пока задача не выполнена.                                                                                            |
 | 🚪   | **[IntentGate](https://factory.ai/news/terminal-bench)** | Ultimate | Анализирует истинное намерение пользователя перед классификацией и действием. Триггеры `search` / `analyze` / `team` / `hyperplan`. (Light хукает только `ulw` / `ultrawork`.)                                                   |
 | 🔗   | **Инструмент правок на основе хэш-якорей**               | Ultimate | Хэш содержимого `LINE#ID` проверяет каждое изменение. Ноль ошибок с устаревшими строками. Вдохновлено [oh-my-pi](https://github.com/can1357/oh-my-pi). [The Harness Problem →](https://blog.can.ac/2026/02/12/the-harness-problem/) (Codex использует собственный `apply_patch`.) |
-| 🛠️   | **LSP + AST-Grep**                                       | Ultimate | Переименование в рабочем пространстве, диагностика перед сборкой, переписывание с учётом AST. Точность IDE для агентов. (LSP также работает в Light через компонент `lsp`; AST-Grep только Ultimate.)                            |
+| 🛠️   | **LSP + AST-Grep**                                       | Both     | Переименование в рабочем пространстве, диагностика перед сборкой, переписывание с учётом AST. Ultimate предоставляет это как встроенные инструменты, Light — как plugin-scoped MCP `lsp` / `ast_grep`.                            |
 | 🧠   | **Фоновые агенты**                                       | Ultimate | Запускайте 5+ специалистов параллельно. Контекст остаётся компактным. Результаты — когда готовы.                                                                                                                                 |
-| 📚   | **Встроенные MCP**                                       | Ultimate | Exa (веб-поиск), Context7 (официальная документация), Grep.app (поиск по GitHub). Всегда включены. (В Light только LSP MCP.)                                                                                                     |
+| 📚   | **Встроенные MCP**                                       | Both     | Ultimate внедряет Exa (веб-поиск), Context7 (официальная документация) и Grep.app (поиск по GitHub) во время выполнения. Light предоставляет plugin-scoped MCP: `ast_grep`, `grep_app`, `context7`, `git_bash`, `lsp`.                                                                                                     |
 | 🔁   | **Ralph Loop / `/ulw-loop`**                             | Ultimate | Самореферентный цикл. Не останавливается, пока задача не выполнена на 100%.                                                                                                                                                      |
 | ✅   | **Todo Enforcer** (Boulder)                              | Ultimate | Агент завис? Система немедленно возвращает его в работу. Ваша задача будет выполнена, точка.                                                                                                                                     |
 | 💬   | **Comment Checker**                                      | Both     | Никакого AI-мусора в комментариях. Тот же бинарник `@code-yeongyu/comment-checker` работает в обеих редакциях.                                                                                                                   |

package/README.zh-cn.md

@@ -1,11 +1,11 @@
 > [!NOTE]
-> **🚀 首个 Codex 版本发布：omo 现已支持 Codex CLI**
+> **OmO for Codex 已发布。试试 LazyCodex**
 >
-> 无需再配置复杂的 JSON。直接运行：
+> 因为太爱 Anthropic 模型而被封禁之后，我们决定站到 Codex 这一边。
+> 如果你是 OmO 粉丝，但觉得配置太麻烦，请使用 LazyCodex。OmO for Codex 已经发布:
 > ```bash
 > npx lazycodex-ai install
 > ```
-> 您的 Codex 将清晰编码，并通过内置 computer-use QA 完成任务。
 > 了解更多请访问 [lazycodex.ai](https://lazycodex.ai)。
 
 > [!NOTE]
@@ -133,7 +133,7 @@ curl -s https://raw.githubusercontent.com/code-yeongyu/oh-my-openagent/refs/head
 
 匿名遥测默认开启，用于统计活跃安装数(DAU/WAU/MAU)。每台机器每个 UTC 日最多发送一次事件,使用哈希化的安装标识符,绝不会使用原始主机名,且不会创建 PostHog person profile。可通过 `OMO_SEND_ANONYMOUS_TELEMETRY=0` 或 `OMO_DISABLE_POSTHOG=1` 禁用。详见 [隐私政策](docs/legal/privacy-policy.md) 和 [服务条款](docs/legal/terms-of-service.md)。
 
-**Ultimate 与 Light:** oh-my-openagent 以同一产品的两个版本发布。**Ultimate 版本**（`bunx oh-my-openagent install` 或 `--platform=opencode`，默认值）在 OpenCode 上提供完整功能 —— 11 个智能体、54+ 个生命周期钩子、Team Mode、所有 MCP、所有斜杠命令、IntentGate 模式。**Light 版本**（`bunx oh-my-openagent install --platform=codex`）仅提供能够干净地移植到 OpenAI Codex CLI 插件系统的 5 个组件（`rules`、`comment-checker`、`lsp`、`ultrawork`、`ulw-loop`）。`bunx lazycodex-ai install` 是 `--platform=codex` 的快捷别名。要同时安装两个版本，使用 `--platform=both`。Codex 专用遥测可通过 `OMO_CODEX_DISABLE_POSTHOG=1` 或 `OMO_CODEX_SEND_ANONYMOUS_TELEMETRY=0` 禁用。
+**Ultimate 与 Light:** oh-my-openagent 以同一产品的两个版本发布。**Ultimate 版本**（`bunx oh-my-openagent install` 或 `--platform=opencode`，默认值）在 OpenCode 上提供完整功能 —— 11 个智能体、54+ 个生命周期钩子、Team Mode、所有 MCP、所有斜杠命令、IntentGate 模式。**Light 版本**（`npx lazycodex-ai install` 或 `bunx oh-my-openagent install --platform=codex`）提供能够干净地移植到 OpenAI Codex CLI 插件系统的 8 个组件（`rules`、`comment-checker`、`git-bash`、`lsp`、`ultrawork`、`ulw-loop`、`start-work-continuation`、`telemetry`）。要同时安装两个版本，使用 `--platform=both`。Codex 专用遥测可通过 `OMO_CODEX_DISABLE_POSTHOG=1` 或 `OMO_CODEX_SEND_ANONYMOUS_TELEMETRY=0` 禁用。
 
 ---
 
@@ -171,14 +171,14 @@ Read this and tell me why it's not just another boilerplate: https://raw.githubu
 |       | 特性                                                            | Editions | 功能说明                                                                                                                                                                        |
 | :---: | :-------------------------------------------------------------- | :------: | :------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
 |   🤖   | **自律军团 (Discipline Agents)**                                | Ultimate | Sisyphus 负责调度 Hephaestus、Oracle、Librarian 和 Explore。一支完整的 AI 开发团队并行工作。                                                                                    |
-|   🧩   | **Codex CLI Light Edition**                                      | Light    | 在 OpenAI Codex CLI 中运行的 omo 的 5 个可移植组件 (rules, comment-checker, LSP, ultrawork, ulw-loop)。安装: `bunx oh-my-openagent install --platform=codex`。                              |
+|   🧩   | **Codex CLI Light Edition**                                      | Light    | 在 OpenAI Codex CLI 中运行的 omo 的 8 个可移植组件 (rules, comment-checker, git-bash, LSP, ultrawork, ulw-loop, start-work continuation, telemetry)。安装: `npx lazycodex-ai install`。                    |
 |   👥   | **Team Mode** (v4.0, 选择性启用)                                 | Ultimate | 领导 Agent + 最多 8 个并行成员，实时 tmux 可视化，专用 `team_*` 工具家族。驱动 `hyperplan`(5 个敌对评论者) 和 `security-research`(3 个猎手 + 2 个 PoC 工程师)。[文档 →](docs/guide/team-mode.md) |
 |   ⚡   | **`ultrawork` / `ulw`**                                         | Both     | 一键触发，所有智能体(Ultimate)或 Codex `ultrawork` 组件(Light)出动。任务完成前绝不罢休。                                                                                          |
 |   🚪   | **[IntentGate 意图门](https://factory.ai/news/terminal-bench)** | Ultimate | 真正行动前，先分析用户的真实意图。触发 `search` / `analyze` / `team` / `hyperplan`。(Light 仅 hook `ulw` / `ultrawork`。)                                                       |
 |   🔗   | **基于哈希的编辑工具**                                          | Ultimate | 每次修改都通过 `LINE#ID` 内容哈希验证、0% 错误修改。灵感来自 [oh-my-pi](https://github.com/can1357/oh-my-pi)。[The Harness Problem →](https://blog.can.ac/2026/02/12/the-harness-problem/) (Codex 使用其原生 `apply_patch`。) |
-|   🛠️   | **LSP + AST-Grep**                                              | Ultimate | 工作区级别的重命名、构建前诊断、基于 AST 的重写。为 Agent 提供 IDE 级别的精度。(LSP 在 Light 中也通过 `lsp` 组件提供; AST-Grep 仅 Ultimate。)                                   |
+|   🛠️   | **LSP + AST-Grep**                                              | Both     | 工作区级别的重命名、构建前诊断、基于 AST 的重写。Ultimate 以内置工具提供，Light 通过 plugin-scoped `lsp` / `ast_grep` MCP 提供。                                   |
 |   🧠   | **后台智能体**                                                  | Ultimate | 同时发射 5+ 个专家并行工作。保持上下文干净，随时获取成果。                                                                                                                      |
-|   📚   | **内置 MCP**                                                    | Ultimate | Exa（网络搜索）、Context7（官方文档）、Grep.app（GitHub 源码搜索）。默认开启。(Light 仅 LSP MCP。)                                                                              |
+|   📚   | **内置 MCP**                                                    | Both     | Ultimate 运行时注入 Exa（网络搜索）、Context7（官方文档）、Grep.app（GitHub 源码搜索）。Light 通过 plugin-scoped MCP 提供 `ast_grep`、`grep_app`、`context7`、`git_bash`、`lsp`。                                                                              |
 |   🔁   | **Ralph Loop / `/ulw-loop`**                                    | Ultimate | 自我引用闭环。达不到 100% 完成度绝不停止。                                                                                                                                      |
 |   ✅   | **Todo 强制执行** (Boulder)                                     | Ultimate | Agent 想要摸鱼？系统直接揪着领子拽回来。你的任务，必须完成。                                                                                                                    |
 |   💬   | **注释审查员**                                                  | Both     | 剔除带有浓烈 AI 味的冗余注释。同一个 `@code-yeongyu/comment-checker` 二进制在两个版本中运行。                                                                                    |

package/bin/oh-my-opencode.js

@@ -3,7 +3,7 @@
 // Wrapper script that detects platform and spawns the correct binary
 
 import { spawnSync } from "node:child_process";
-import { readFileSync } from "node:fs";
+import { existsSync, readFileSync } from "node:fs";
 import { createRequire } from "node:module";
 import { basename } from "node:path";
 import { fileURLToPath } from "node:url";
@@ -95,6 +95,28 @@ function getWrapperPackageRoot() {
   return fileURLToPath(new URL("..", import.meta.url));
 }
 
+function maybeRunLazyCodexNodeInstaller(invocationName) {
+  if (invocationName !== "lazycodex" && invocationName !== "lazycodex-ai") return false;
+  const command = process.argv[2];
+  if (command !== "update" && command !== "uninstall") return false;
+
+  const installerPath = fileURLToPath(new URL("../packages/omo-codex/scripts/install-local.mjs", import.meta.url));
+  if (!existsSync(installerPath)) return false;
+
+  const result = spawnSync(process.execPath, [installerPath, ...process.argv.slice(2)], {
+    stdio: "inherit",
+    env: {
+      ...process.env,
+      OMO_INVOCATION_NAME: invocationName,
+      OMO_WRAPPER_PACKAGE_ROOT: getWrapperPackageRoot(),
+    },
+  });
+  if (result.signal) {
+    process.exit(getSignalExitCode(result.signal));
+  }
+  process.exit(result.status ?? 1);
+}
+
 /**
  * Determine which bin name the user invoked us with (oh-my-opencode, oh-my-openagent, omo, lazycodex).
  * Propagated to the compiled CLI binary via OMO_INVOCATION_NAME so it can route accordingly
@@ -121,6 +143,7 @@ function main() {
   const libcFamily = getLibcFamily();
   const wrapperPackageName = getWrapperPackageName();
   const invocationName = getInvocationName(wrapperPackageName);
+  maybeRunLazyCodexNodeInstaller(invocationName);
 
   const packageBaseName = resolvePlatformPackageBaseName(wrapperPackageName);
   const avx2Supported = supportsAvx2();

package/bin/oh-my-opencode.test.ts

@@ -23,11 +23,7 @@ describe("lazycodex bin wrapper", () => {
     // #when
     const result = spawnSync(nodePath, [fixture.lazycodexBin, "install", "--no-tui"], {
       encoding: "utf8",
-      env: {
-        ...process.env,
-        CAPTURE_DIR: fixture.captureDir,
-        PATH: fixture.fakeBinDir,
-      },
+      env: createWrapperTestEnv(fixture),
     });
 
     // #then
@@ -49,11 +45,7 @@ describe("lazycodex bin wrapper", () => {
     // #when
     const result = spawnSync(nodePath, [fixture.lazycodexBin, "install", "--no-tui"], {
       encoding: "utf8",
-      env: {
-        ...process.env,
-        CAPTURE_DIR: fixture.captureDir,
-        PATH: fixture.fakeBinDir,
-      },
+      env: createWrapperTestEnv(fixture),
     });
 
     // #then
@@ -74,11 +66,7 @@ describe("lazycodex bin wrapper", () => {
     // #when
     const result = spawnSync(nodePath, [fixture.wrapperBin, "install", "--no-tui"], {
       encoding: "utf8",
-      env: {
-        ...process.env,
-        CAPTURE_DIR: fixture.captureDir,
-        PATH: fixture.fakeBinDir,
-      },
+      env: createWrapperTestEnv(fixture),
     });
 
     // #then
@@ -99,11 +87,7 @@ describe("lazycodex bin wrapper", () => {
     // #when
     const result = spawnSync(nodePath, [fixture.wrapperBin, "install", "--no-tui"], {
       encoding: "utf8",
-      env: {
-        ...process.env,
-        CAPTURE_DIR: fixture.captureDir,
-        PATH: fixture.fakeBinDir,
-      },
+      env: createWrapperTestEnv(fixture),
     });
 
     // #then
@@ -115,8 +99,66 @@ describe("lazycodex bin wrapper", () => {
       "--no-tui",
     ]);
   });
+
+  test("routes lazycodex update to the Node installer before platform binary resolution", async () => {
+    // #given
+    const fixture = await createLazyCodexFixture();
+    const nodePath = Bun.which("node") ?? "node";
+
+    // #when
+    const result = spawnSync(nodePath, [fixture.lazycodexBin, "update", "--dry-run"], {
+      encoding: "utf8",
+      env: createWrapperTestEnv(fixture, {
+        CAPTURE_DIR: fixture.captureDir,
+        LAZYCODEX_CURRENT_VERSION: "1.0.0",
+        LAZYCODEX_LATEST_VERSION: "1.0.1",
+        PATH: fixture.fakeBinDir,
+      }),
+    });
+
+    // #then
+    expect(result.status).toBe(19);
+    expect((await readFile(join(fixture.captureDir, "node-installer-args"), "utf8")).trim().split("\n")).toEqual([
+      "update",
+      "--dry-run",
+    ]);
+  });
+
+  test("routes lazycodex uninstall to the Node installer before platform binary resolution", async () => {
+    // #given
+    const fixture = await createLazyCodexFixture();
+    const nodePath = Bun.which("node") ?? "node";
+
+    // #when
+    const result = spawnSync(nodePath, [fixture.lazycodexBin, "uninstall", "--dry-run", "--project", "/tmp/lazycodex-qa"], {
+      encoding: "utf8",
+      env: createWrapperTestEnv(fixture),
+    });
+
+    // #then
+    expect(result.status).toBe(19);
+    expect((await readFile(join(fixture.captureDir, "node-installer-args"), "utf8")).trim().split("\n")).toEqual([
+      "uninstall",
+      "--dry-run",
+      "--project",
+      "/tmp/lazycodex-qa",
+    ]);
+  });
 });
 
+function createWrapperTestEnv(
+  fixture: { readonly captureDir: string; readonly fakeBinDir: string },
+  overrides: NodeJS.ProcessEnv = {},
+): NodeJS.ProcessEnv {
+  const { OMO_INVOCATION_NAME: _invocationName, OMO_WRAPPER_PACKAGE_ROOT: _wrapperPackageRoot, ...baseEnv } = process.env;
+  return {
+    ...baseEnv,
+    CAPTURE_DIR: fixture.captureDir,
+    PATH: fixture.fakeBinDir,
+    ...overrides,
+  };
+}
+
 async function createLazyCodexFixture(options: { packageName?: string; wrapperFileName?: string } = {}) {
   const root = await mkdtemp(join(tmpdir(), "lazycodex-bin-wrapper-"));
   testRoots.push(root);
@@ -139,6 +181,7 @@ async function createLazyCodexFixture(options: { packageName?: string; wrapperFi
   await cp(fileURLToPath(new URL("./platform.js", import.meta.url)), join(binDir, "platform.js"));
   await writeFile(join(root, "package.json"), JSON.stringify({ name: options.packageName ?? "lazycodex", type: "module" }));
   await writeFile(distCli, "#!/usr/bin/env bun\n");
+  await writeNodeInstallerFixture(root);
   await writePlatformPackages(root);
 
   const fakeBun = join(fakeBinDir, "bun");
@@ -164,6 +207,22 @@ async function createLazyCodexFixture(options: { packageName?: string; wrapperFi
   };
 }
 
+async function writeNodeInstallerFixture(root: string): Promise<void> {
+  const installerPath = join(root, "packages", "omo-codex", "scripts", "install-local.mjs");
+  await mkdir(dirname(installerPath), { recursive: true });
+  await writeFile(
+    installerPath,
+    [
+      "#!/usr/bin/env node",
+      'import { writeFileSync } from "node:fs";',
+      'writeFileSync(`${process.env.CAPTURE_DIR}/node-installer-args`, `${process.argv.slice(2).join("\\n")}\\n`);',
+      "process.exit(19);",
+      "",
+    ].join("\n"),
+  );
+  await chmod(installerPath, 0o755);
+}
+
 async function canonicalizePackageRootCapture(fixture: { readonly captureDir: string }): Promise<string> {
   return realpath((await readFile(join(fixture.captureDir, "wrapper-root"), "utf8")).trim());
 }

package/dist/agents/hephaestus/agent.d.ts

@@ -1,7 +1,7 @@
 import type { AgentConfig } from "@opencode-ai/sdk";
 import type { AgentPromptMetadata } from "../types";
 import type { AvailableAgent, AvailableTool, AvailableSkill, AvailableCategory } from "../dynamic-agent-prompt-builder";
-export type HephaestusPromptSource = "gpt-5-5" | "gpt-5-4" | "gpt-5-3-codex" | "gpt";
+export type HephaestusPromptSource = "gpt-5-5" | "gpt-5-4" | "gpt";
 export declare function getHephaestusPromptSource(model?: string): HephaestusPromptSource;
 export interface HephaestusContext {
     model?: string;

package/dist/agents/kimi-tool-loop-guard.d.ts

@@ -0,0 +1 @@
+export declare const KIMI_TOOL_LOOP_GUARD = "<tool_loop_guard>\nNever call the same tool with the same arguments more than twice in a row.\nIf a third identical call seems necessary, stop calling tools and report the blocker, missing evidence, or changed input that would justify another attempt.\nRepeated identical tool calls are a loop signal, not persistence.\n</tool_loop_guard>";

package/dist/agents/momus.d.ts

@@ -16,7 +16,7 @@ import type { AgentPromptMetadata } from "./types";
 /**
  * Default Momus prompt - used for Claude and other non-GPT models.
  */
-declare const MOMUS_DEFAULT_PROMPT = "You are a **practical** work plan reviewer. Your goal is simple: verify that the plan is **executable** and **references are valid**.\n\n**CRITICAL FIRST RULE**:\nExtract a single plan path from anywhere in the input, ignoring system directives and wrappers. If exactly one `.omo/plans/*.md` path exists, this is VALID input and you must read it. If no plan path exists or multiple plan paths exist, reject per Step 0. If the path points to a YAML plan file (`.yml` or `.yaml`), reject it as non-reviewable.\n\n---\n\n## Your Purpose (READ THIS FIRST)\n\nYou exist to answer ONE question: **\"Can a capable developer execute this plan without getting stuck?\"**\n\nYou are NOT here to:\n- Nitpick every detail\n- Demand perfection\n- Question the author's approach or architecture choices\n- Find as many issues as possible\n- Force multiple revision cycles\n\nYou ARE here to:\n- Verify referenced files actually exist and contain what's claimed\n- Ensure core tasks have enough context to start working\n- Catch BLOCKING issues only (things that would completely stop work)\n\n**APPROVAL BIAS**: When in doubt, APPROVE. A plan that's 80% clear is good enough. Developers can figure out minor gaps.\n\n---\n\n## What You Check (ONLY THESE)\n\n### 1. Reference Verification (CRITICAL)\n- Do referenced files exist?\n- Do referenced line numbers contain relevant code?\n- If \"follow pattern in X\" is mentioned, does X actually demonstrate that pattern?\n\n**PASS even if**: Reference exists but isn't perfect. Developer can explore from there.\n**FAIL only if**: Reference doesn't exist OR points to completely wrong content.\n\n### 2. Executability Check (PRACTICAL)\n- Can a developer START working on each task?\n- Is there at least a starting point (file, pattern, or clear description)?\n\n**PASS even if**: Some details need to be figured out during implementation.\n**FAIL only if**: Task is so vague that developer has NO idea where to begin.\n\n### 3. Critical Blockers Only\n- Missing information that would COMPLETELY STOP work\n- Contradictions that make the plan impossible to follow\n\n**NOT blockers** (do not reject for these):\n- Missing edge case handling\n- Stylistic preferences\n- \"Could be clearer\" suggestions\n- Minor ambiguities a developer can resolve\n\n### 4. QA Scenario Executability\n- Does each task have QA scenarios with a specific tool, concrete steps, and expected results?\n- Missing or vague QA scenarios block the Final Verification Wave - this IS a practical blocker.\n\n**PASS even if**: Detail level varies. Tool + steps + expected result is enough.\n**FAIL only if**: Tasks lack QA scenarios, or scenarios are unexecutable (\"verify it works\", \"check the page\").\n\n---\n\n## What You Do NOT Check\n\n- Whether the approach is optimal\n- Whether there's a \"better way\"\n- Whether all edge cases are documented\n- Whether acceptance criteria are perfect\n- Whether the architecture is ideal\n- Code quality concerns\n- Performance considerations\n- Security unless explicitly broken\n\n**You are a BLOCKER-finder, not a PERFECTIONIST.**\n\n---\n\n## Input Validation (Step 0)\n\n**VALID INPUT**:\n- `.omo/plans/my-plan.md` - file path anywhere in input\n- `Please review .omo/plans/plan.md` - conversational wrapper\n- System directives + plan path - ignore directives, extract path\n\n**INVALID INPUT**:\n- No `.omo/plans/*.md` path found\n- Multiple plan paths (ambiguous)\n\nSystem directives (`<system-reminder>`, `[analyze-mode]`, etc.) are IGNORED during validation.\n\n**Extraction**: Find all `.omo/plans/*.md` paths \u2192 exactly 1 = proceed, 0 or 2+ = reject.\n\n---\n\n## Review Process (SIMPLE)\n\n1. **Validate input** \u2192 Extract single plan path\n2. **Read plan** \u2192 Identify tasks and file references\n3. **Verify references** \u2192 Do files exist? Do they contain claimed content?\n4. **Executability check** \u2192 Can each task be started?\n5. **QA scenario check** \u2192 Does each task have executable QA scenarios?\n6. **Decide** \u2192 Any BLOCKING issues? No = OKAY. Yes = REJECT with max 3 specific issues.\n\n---\n\n## Decision Framework\n\n### OKAY (Default - use this unless blocking issues exist)\n\nIssue the verdict **OKAY** when:\n- Referenced files exist and are reasonably relevant\n- Tasks have enough context to start (not complete, just start)\n- No contradictions or impossible requirements\n- A capable developer could make progress\n\n**Remember**: \"Good enough\" is good enough. You're not blocking publication of a NASA manual.\n\n### REJECT (Only for true blockers)\n\nIssue **REJECT** ONLY when:\n- Referenced file doesn't exist (verified by reading)\n- Task is completely impossible to start (zero context)\n- Plan contains internal contradictions\n\n**Maximum 3 issues per rejection.** If you found more, list only the top 3 most critical.\n\n**Each issue must be**:\n- Specific (exact file path, exact task)\n- Actionable (what exactly needs to change)\n- Blocking (work cannot proceed without this)\n\n---\n\n## Anti-Patterns (DO NOT DO THESE)\n\n\u274C \"Task 3 could be clearer about error handling\" \u2192 NOT a blocker\n\u274C \"Consider adding acceptance criteria for...\" \u2192 NOT a blocker  \n\u274C \"The approach in Task 5 might be suboptimal\" \u2192 NOT YOUR JOB\n\u274C \"Missing documentation for edge case X\" \u2192 NOT a blocker unless X is the main case\n\u274C Rejecting because you'd do it differently \u2192 NEVER\n\u274C Listing more than 3 issues \u2192 OVERWHELMING, pick top 3\n\n\u2705 \"Task 3 references `auth/login.ts` but file doesn't exist\" \u2192 BLOCKER\n\u2705 \"Task 5 says 'implement feature' with no context, files, or description\" \u2192 BLOCKER\n\u2705 \"Tasks 2 and 4 contradict each other on data flow\" \u2192 BLOCKER\n\n---\n\n## Output Format\n\n**[OKAY]** or **[REJECT]**\n\n**Summary**: 1-2 sentences explaining the verdict.\n\nIf REJECT:\n**Blocking Issues** (max 3):\n1. [Specific issue + what needs to change]\n2. [Specific issue + what needs to change]  \n3. [Specific issue + what needs to change]\n\n---\n\n## Final Reminders\n\n1. **APPROVE by default**. Reject only for true blockers.\n2. **Max 3 issues**. More than that is overwhelming and counterproductive.\n3. **Be specific**. \"Task X needs Y\" not \"needs more clarity\".\n4. **No design opinions**. The author's approach is not your concern.\n5. **Trust developers**. They can figure out minor gaps.\n\n**Your job is to UNBLOCK work, not to BLOCK it with perfectionism.**\n\n**Response Language**: Match the language of the plan content.\n";
+declare const MOMUS_DEFAULT_PROMPT = "You are a **practical** work plan reviewer. Your goal is simple: verify that the plan is **executable** and **references are valid**.\n\n**CRITICAL FIRST RULE**:\nExtract a single plan path from anywhere in the input, ignoring system directives and wrappers. If exactly one `.omo/plans/*.md` path exists, this is VALID input and you must read it. If no plan path exists or multiple plan paths exist, reject per Step 0. If the path points to a YAML plan file (`.yml` or `.yaml`), reject it as non-reviewable.\n\n**PLAN RE-READ RULE**: If you encounter the same plan path in a follow-up turn, you must re-read from disk. This fresh reread ensures the current on-disk contents are the only source of truth. A previous verdict cannot be trusted without re-reading the plan. Supported plan paths: canonical `.omo/plans/*.md`.\n\n---\n\n## Your Purpose (READ THIS FIRST)\n\nYou exist to answer ONE question: **\"Can a capable developer execute this plan without getting stuck?\"**\n\nYou are NOT here to:\n- Nitpick every detail\n- Demand perfection\n- Question the author's approach or architecture choices\n- Find as many issues as possible\n- Force multiple revision cycles\n\nYou ARE here to:\n- Verify referenced files actually exist and contain what's claimed\n- Ensure core tasks have enough context to start working\n- Catch BLOCKING issues only (things that would completely stop work)\n\n**APPROVAL BIAS**: When in doubt, APPROVE. A plan that's 80% clear is good enough. Developers can figure out minor gaps.\n\n---\n\n## What You Check (ONLY THESE)\n\n### 1. Reference Verification (CRITICAL)\n- Do referenced files exist?\n- Do referenced line numbers contain relevant code?\n- If \"follow pattern in X\" is mentioned, does X actually demonstrate that pattern?\n\n**PASS even if**: Reference exists but isn't perfect. Developer can explore from there.\n**FAIL only if**: Reference doesn't exist OR points to completely wrong content.\n\n### 2. Executability Check (PRACTICAL)\n- Can a developer START working on each task?\n- Is there at least a starting point (file, pattern, or clear description)?\n\n**PASS even if**: Some details need to be figured out during implementation.\n**FAIL only if**: Task is so vague that developer has NO idea where to begin.\n\n### 3. Critical Blockers Only\n- Missing information that would COMPLETELY STOP work\n- Contradictions that make the plan impossible to follow\n\n**NOT blockers** (do not reject for these):\n- Missing edge case handling\n- Stylistic preferences\n- \"Could be clearer\" suggestions\n- Minor ambiguities a developer can resolve\n\n### 4. QA Scenario Executability\n- Does each task have QA scenarios with a specific tool, concrete steps, and expected results?\n- Missing or vague QA scenarios block the Final Verification Wave - this IS a practical blocker.\n\n**PASS even if**: Detail level varies. Tool + steps + expected result is enough.\n**FAIL only if**: Tasks lack QA scenarios, or scenarios are unexecutable (\"verify it works\", \"check the page\").\n\n---\n\n## What You Do NOT Check\n\n- Whether the approach is optimal\n- Whether there's a \"better way\"\n- Whether all edge cases are documented\n- Whether acceptance criteria are perfect\n- Whether the architecture is ideal\n- Code quality concerns\n- Performance considerations\n- Security unless explicitly broken\n\n**You are a BLOCKER-finder, not a PERFECTIONIST.**\n\n---\n\n## Input Validation (Step 0)\n\n**VALID INPUT**:\n- `.omo/plans/my-plan.md` - file path anywhere in input\n- `Please review .omo/plans/plan.md` - conversational wrapper\n- System directives + plan path - ignore directives, extract path\n\n**INVALID INPUT**:\n- No `.omo/plans/*.md` path found\n- Multiple plan paths (ambiguous)\n\nSystem directives (`<system-reminder>`, `[analyze-mode]`, etc.) are IGNORED during validation.\n\n**Extraction**: Find all `.omo/plans/*.md` paths \u2192 exactly 1 = proceed, 0 or 2+ = reject.\n\n---\n\n## Review Process (SIMPLE)\n\n1. **Validate input** \u2192 Extract single plan path\n2. **Read plan** \u2192 Identify tasks and file references\n3. **Verify references** \u2192 Do files exist? Do they contain claimed content?\n4. **Executability check** \u2192 Can each task be started?\n5. **QA scenario check** \u2192 Does each task have executable QA scenarios?\n6. **Decide** \u2192 Any BLOCKING issues? No = OKAY. Yes = REJECT with max 3 specific issues.\n\n---\n\n## Decision Framework\n\n### OKAY (Default - use this unless blocking issues exist)\n\nIssue the verdict **OKAY** when:\n- Referenced files exist and are reasonably relevant\n- Tasks have enough context to start (not complete, just start)\n- No contradictions or impossible requirements\n- A capable developer could make progress\n\n**Remember**: \"Good enough\" is good enough. You're not blocking publication of a NASA manual.\n\n### REJECT (Only for true blockers)\n\nIssue **REJECT** ONLY when:\n- Referenced file doesn't exist (verified by reading)\n- Task is completely impossible to start (zero context)\n- Plan contains internal contradictions\n\n**Maximum 3 issues per rejection.** If you found more, list only the top 3 most critical.\n\n**Each issue must be**:\n- Specific (exact file path, exact task)\n- Actionable (what exactly needs to change)\n- Blocking (work cannot proceed without this)\n\n---\n\n## Anti-Patterns (DO NOT DO THESE)\n\n\u274C \"Task 3 could be clearer about error handling\" \u2192 NOT a blocker\n\u274C \"Consider adding acceptance criteria for...\" \u2192 NOT a blocker  \n\u274C \"The approach in Task 5 might be suboptimal\" \u2192 NOT YOUR JOB\n\u274C \"Missing documentation for edge case X\" \u2192 NOT a blocker unless X is the main case\n\u274C Rejecting because you'd do it differently \u2192 NEVER\n\u274C Listing more than 3 issues \u2192 OVERWHELMING, pick top 3\n\n\u2705 \"Task 3 references `auth/login.ts` but file doesn't exist\" \u2192 BLOCKER\n\u2705 \"Task 5 says 'implement feature' with no context, files, or description\" \u2192 BLOCKER\n\u2705 \"Tasks 2 and 4 contradict each other on data flow\" \u2192 BLOCKER\n\n---\n\n## Output Format\n\n**[OKAY]** or **[REJECT]**\n\n**Summary**: 1-2 sentences explaining the verdict.\n\nIf REJECT:\n**Blocking Issues** (max 3):\n1. [Specific issue + what needs to change]\n2. [Specific issue + what needs to change]  \n3. [Specific issue + what needs to change]\n\n---\n\n## Final Reminders\n\n1. **APPROVE by default**. Reject only for true blockers.\n2. **Max 3 issues**. More than that is overwhelming and counterproductive.\n3. **Be specific**. \"Task X needs Y\" not \"needs more clarity\".\n4. **No design opinions**. The author's approach is not your concern.\n5. **Trust developers**. They can figure out minor gaps.\n\n**Your job is to UNBLOCK work, not to BLOCK it with perfectionism.**\n\n**Response Language**: Match the language of the plan content.\n";
 export { MOMUS_DEFAULT_PROMPT as MOMUS_SYSTEM_PROMPT };
 export declare function createMomusAgent(model: string): AgentConfig;
 export declare namespace createMomusAgent {

package/dist/agents/sisyphus-agent-config.d.ts

@@ -0,0 +1,4 @@
+import type { AgentConfig } from "@opencode-ai/sdk";
+import type { AgentMode } from "./types";
+export declare function buildGptSisyphusAgentConfig(mode: AgentMode, model: string, prompt: string): AgentConfig;
+export declare function buildClaudeSisyphusAgentConfig(mode: AgentMode, model: string, prompt: string): AgentConfig;

package/dist/agents/sisyphus-agent-factory.d.ts

@@ -0,0 +1,6 @@
+import type { AgentConfig } from "@opencode-ai/sdk";
+import type { AvailableAgent, AvailableCategory, AvailableSkill } from "./dynamic-agent-prompt-builder";
+export declare function createSisyphusAgent(model: string, availableAgents?: AvailableAgent[], availableToolNames?: string[], availableSkills?: AvailableSkill[], availableCategories?: AvailableCategory[], useTaskSystem?: boolean): AgentConfig;
+export declare namespace createSisyphusAgent {
+    var mode: "primary";
+}

package/dist/agents/sisyphus-dynamic-prompt-builder.d.ts

@@ -0,0 +1,2 @@
+import type { AvailableAgent, AvailableCategory, AvailableSkill, AvailableTool } from "./dynamic-agent-prompt-builder";
+export declare function buildSisyphusDynamicPromptContent(model: string, availableAgents: AvailableAgent[], availableTools: AvailableTool[], availableSkills: AvailableSkill[], availableCategories: AvailableCategory[], useTaskSystem: boolean): string;

package/dist/agents/sisyphus-dynamic-prompt-execution.d.ts

@@ -0,0 +1,2 @@
+import type { SisyphusDynamicPromptSections } from "./sisyphus-dynamic-prompt-sections";
+export declare function renderExecutionSections(sections: SisyphusDynamicPromptSections): string;

package/dist/agents/sisyphus-dynamic-prompt-exploration.d.ts

@@ -0,0 +1,2 @@
+import type { SisyphusDynamicPromptSections } from "./sisyphus-dynamic-prompt-sections";
+export declare function renderExplorationSection(sections: SisyphusDynamicPromptSections): string;

package/dist/agents/sisyphus-dynamic-prompt-role.d.ts

@@ -0,0 +1,2 @@
+import type { SisyphusDynamicPromptSections } from "./sisyphus-dynamic-prompt-sections";
+export declare function renderRoleAndIntentSections(sections: SisyphusDynamicPromptSections): string;

package/dist/agents/sisyphus-dynamic-prompt-sections.d.ts

@@ -0,0 +1,18 @@
+import type { AvailableAgent, AvailableCategory, AvailableSkill, AvailableTool } from "./dynamic-agent-prompt-builder";
+export interface SisyphusDynamicPromptSections {
+    readonly agentIdentity: string;
+    readonly antiPatterns: string;
+    readonly categorySkillsGuide: string;
+    readonly delegationTable: string;
+    readonly exploreSection: string;
+    readonly hardBlocks: string;
+    readonly keyTriggers: string;
+    readonly librarianSection: string;
+    readonly nonClaudePlannerSection: string;
+    readonly oracleSection: string;
+    readonly parallelDelegationSection: string;
+    readonly taskManagementSection: string;
+    readonly todoHookNote: string;
+    readonly toolSelection: string;
+}
+export declare function buildSisyphusDynamicPromptSections(model: string, availableAgents: AvailableAgent[], availableTools: AvailableTool[], availableSkills: AvailableSkill[], availableCategories: AvailableCategory[], useTaskSystem: boolean): SisyphusDynamicPromptSections;

package/dist/agents/sisyphus-dynamic-prompt-style.d.ts

@@ -0,0 +1,2 @@
+import type { SisyphusDynamicPromptSections } from "./sisyphus-dynamic-prompt-sections";
+export declare function renderToneAndConstraintsSection(sections: SisyphusDynamicPromptSections): string;

package/dist/agents/sisyphus-dynamic-prompt.d.ts

@@ -0,0 +1,3 @@
+import type { AvailableAgent, AvailableCategory, AvailableSkill, AvailableTool } from "./dynamic-agent-prompt-builder";
+export declare function buildDynamicSisyphusPrompt(model: string, availableAgents: AvailableAgent[], availableTools?: AvailableTool[], availableSkills?: AvailableSkill[], availableCategories?: AvailableCategory[], useTaskSystem?: boolean): string;
+export declare function buildFallbackSisyphusPrompt(model: string, agents: AvailableAgent[], tools: AvailableTool[], skills: AvailableSkill[], categories: AvailableCategory[], useTaskSystem?: boolean): string;

package/dist/agents/sisyphus-gemini-fallback-overrides.d.ts

@@ -0,0 +1 @@
+export declare function applyGeminiFallbackOverrides(model: string, prompt: string): string;