feat(agents): add Gemini-optimized prompts for Sisyphus, Sisyphus-Junior, Prometheus, Atlas

Gemini models are aggressively optimistic and avoid tool calls in favor of internal reasoning. These prompts counter that with: - TOOL_CALL_MANDATE sections forcing actual tool usage - Anti-optimism checkpoints before claiming completion - Stronger delegation enforcement (Gemini prefers doing work itself) - Aggressive verification language (subagent results are 'EXTREMELY SUSPICIOUS') - Mandatory thinking checkpoints in Prometheus (prevents jumping to conclusions) - Scope discipline reminders (creativity → implementation quality, not scope creep)
2026-02-22 15:08:24 +09:00
239 changed files with 4368 additions and 7088 deletions
--- a/.issue-comment-2064.md
+++ b/.issue-comment-2064.md
@@ -1,61 +0,0 @@
-[sisyphus-bot] 
-
-## Confirmed Bug
-
-We have identified the root cause of this issue. The bug is in the config writing logic during installation.
-
-### Root Cause
-
-**File:** `src/cli/config-manager/write-omo-config.ts` (line 46)
-
-```typescript
-const merged = deepMergeRecord(existing, newConfig)
-```
-
-When a user runs `oh-my-opencode install` (even just to update settings), the installer:
-1. Reads the existing config (with user's custom model settings)
-2. Generates a **new** config based on detected provider availability
-3. Calls `deepMergeRecord(existing, newConfig)` 
-4. Writes the result back
-
-**The problem:** `deepMergeRecord` overwrites values in `existing` with values from `newConfig`. This means your custom `"model": "openai/gpt-5.2-codex"` gets overwritten by the generated default model (e.g., `anthropic/claude-opus-4-6` if Claude is available).
-
-### Why This Happens
-
-Looking at `deepMergeRecord` (line 24-25):
-```typescript
-} else if (sourceValue !== undefined) {
-  result[key] = sourceValue as TTarget[keyof TTarget]
-}
-```
-
-Any defined value in the source (generated config) overwrites the target (user's config).
-
-### Fix Approach
-
-The merge direction should be reversed to respect user overrides:
-```typescript
-const merged = deepMergeRecord(newConfig, existing)
-```
-
-This ensures:
- User's explicit settings take precedence
- Only new/undefined keys get populated from generated defaults
- Custom model choices are preserved
-
-### SEVERITY: HIGH
-
- **Impact:** User configuration is overwritten without consent
- **Affected Files:** 
-  - `src/cli/config-manager/write-omo-config.ts`
-  - `src/cli/config-manager/deep-merge-record.ts`
- **Trigger:** Running `oh-my-opencode install` (even for unrelated updates)
-
-### Workaround (Until Fix)
-
-Backup your config before running install:
-```bash
-cp ~/.config/opencode/oh-my-opencode.jsonc ~/.config/opencode/oh-my-opencode.jsonc.backup
-```
-
-We're working on a fix that will preserve your explicit model configurations.
--- a/AGENTS.md
+++ b/AGENTS.md
@@ -1,10 +1,10 @@
 # oh-my-opencode — OpenCode Plugin

-**Generated:** 2026-02-24 | **Commit:** fcb90d92 | **Branch:** dev
+**Generated:** 2026-02-21 | **Commit:** 86e3c7d1 | **Branch:** dev

 ## OVERVIEW

-OpenCode plugin (npm: `oh-my-opencode`) that extends Claude Code (OpenCode fork) with multi-agent orchestration, 46 lifecycle hooks, 26 tools, skill/command/MCP systems, and Claude Code compatibility. 1208 TypeScript files, 143k LOC.
+OpenCode plugin (npm: `oh-my-opencode`) that extends Claude Code (OpenCode fork) with multi-agent orchestration, 44 lifecycle hooks, 26 tools, skill/command/MCP systems, and Claude Code compatibility. 1208 TypeScript files, 143k LOC.

 ## STRUCTURE

@@ -14,14 +14,14 @@ oh-my-opencode/
 │   ├── index.ts              # Plugin entry: loadConfig → createManagers → createTools → createHooks → createPluginInterface
 │   ├── plugin-config.ts      # JSONC multi-level config: user → project → defaults (Zod v4)
 │   ├── agents/               # 11 agents (Sisyphus, Hephaestus, Oracle, Librarian, Explore, Atlas, Prometheus, Metis, Momus, Multimodal-Looker, Sisyphus-Junior)
-| `hooks/`                # 46 hooks across 39 directories + 6 standalone files
+│   ├── hooks/                # 44 hooks across 39 directories + 6 standalone files
 │   ├── tools/                # 26 tools across 15 directories
 │   ├── features/             # 19 feature modules (background-agent, skill-loader, tmux, MCP-OAuth, etc.)
 │   ├── shared/               # 100+ utility files in 13 categories
 │   ├── config/               # Zod v4 schema system (22+ files)
 │   ├── cli/                  # CLI: install, run, doctor, mcp-oauth (Commander.js)
 │   ├── mcp/                  # 3 built-in remote MCPs (websearch, context7, grep_app)
-│   ├── plugin/               # 8 OpenCode hook handlers + 46 hook composition
+│   ├── plugin/               # 8 OpenCode hook handlers + 44 hook composition
 │   └── plugin-handlers/      # 6-phase config loading pipeline
 ├── packages/                 # Monorepo: comment-checker, opencode-sdk, 10 platform binaries
 └── local-ignore/             # Dev-only test fixtures
@@ -34,7 +34,7 @@ OhMyOpenCodePlugin(ctx)
  ├─→ loadPluginConfig()         # JSONC parse → project/user merge → Zod validate → migrate
  ├─→ createManagers()           # TmuxSessionManager, BackgroundManager, SkillMcpManager, ConfigHandler
  ├─→ createTools()              # SkillContext + AvailableCategories + ToolRegistry (26 tools)
-  ├─→ createHooks()              # 3-tier: Core(37) + Continuation(7) + Skill(2) = 46 hooks
+  ├─→ createHooks()              # 3-tier: Core(35) + Continuation(7) + Skill(2) = 44 hooks
  └─→ createPluginInterface()    # 8 OpenCode hook handlers → PluginInterface
 ```

@@ -87,7 +87,7 @@ Fields: agents (14 overridable, 21 fields each), categories (8 built-in + custom

 - **Test pattern**: Bun test (`bun:test`), co-located `*.test.ts`, given/when/then style (nested describe with `#given`/`#when`/`#then` prefixes)
 - **Factory pattern**: `createXXX()` for all tools, hooks, agents
- **Hook tiers**: Session (23) → Tool-Guard (10) → Transform (4) → Continuation (7) → Skill (2)
+- **Hook tiers**: Session (22) → Tool-Guard (10) → Transform (4) → Continuation (7) → Skill (2)
 - **Agent modes**: `primary` (respects UI model) vs `subagent` (own fallback chain) vs `all`
 - **Model resolution**: 3-step: override → category-default → provider-fallback → system-default
 - **Config format**: JSONC with comments, Zod v4 validation, snake_case keys
--- a/README.ja.md
+++ b/README.ja.md
@@ -217,9 +217,9 @@ MCPサーバーがあなたのコンテキスト予算を食いつぶしてい
 [oh-my-pi](https://github.com/can1357/oh-my-pi) に触発され、**Hashline**を実装しました。エージェントが読むすべての行にコンテンツハッシュがタグ付けされて返されます：

 ```
-11#VK| function hello() {
-22#XJ|   return "world";
-33#MB| }
+11#VK: function hello() {
+22#XJ:   return "world";
+33#MB: }
 ```

 エージェントはこのタグを参照して編集します。最後に読んだ後でファイルが変更されていた場合、ハッシュが一致せず、コードが壊れる前に編集が拒否されます。空白を正確に再現する必要もなく、間違った行を編集するエラー (stale-line) もありません。
--- a/README.ko.md
+++ b/README.ko.md
@@ -216,9 +216,9 @@ MCP 서버들이 당신의 컨텍스트 예산을 다 잡아먹죠. 우리가
 [oh-my-pi](https://github.com/can1357/oh-my-pi)에서 영감을 받아, **Hashline**을 구현했습니다. 에이전트가 읽는 모든 줄에는 콘텐츠 해시 태그가 붙어 나옵니다:

 ```
-11#VK| function hello() {
-22#XJ|   return "world";
-33#MB| }
+11#VK: function hello() {
+22#XJ:   return "world";
+33#MB: }
 ```

 에이전트는 이 태그를 참조해서 편집합니다. 마지막으로 읽은 후 파일이 변경되었다면 해시가 일치하지 않아 코드가 망가지기 전에 편집이 거부됩니다. 공백을 똑같이 재현할 필요도 없고, 엉뚱한 줄을 수정하는 에러(stale-line)도 없습니다.
--- a/README.md
+++ b/README.md
@@ -220,9 +220,9 @@ The harness problem is real. Most agent failures aren't the model. It's the edit
 Inspired by [oh-my-pi](https://github.com/can1357/oh-my-pi), we implemented **Hashline**. Every line the agent reads comes back tagged with a content hash:

 ```
-11#VK| function hello() {
-22#XJ|   return "world";
-33#MB| }
+11#VK: function hello() {
+22#XJ:   return "world";
+33#MB: }
 ```

 The agent edits by referencing those tags. If the file changed since the last read, the hash won't match and the edit is rejected before corruption. No whitespace reproduction. No stale-line errors.
--- a/README.zh-cn.md
+++ b/README.zh-cn.md
@@ -218,9 +218,9 @@ Harness 问题是真的。绝大多数所谓的 Agent 故障，其实并不是
 受 [oh-my-pi](https://github.com/can1357/oh-my-pi) 的启发，我们实现了 **Hashline** 技术。Agent 读到的每一行代码，末尾都会打上一个强绑定的内容哈希值：

 ```
-11#VK| function hello() {
-22#XJ|   return "world";
-33#MB| }
+11#VK: function hello() {
+22#XJ:   return "world";
+33#MB: }
 ```

 Agent 发起修改时，必须通过这些标签引用目标行。如果在此期间文件发生过变化，哈希验证就会失败，从而在代码被污染前直接驳回。不再有缩进空格错乱，彻底告别改错行的惨剧。
--- a/assets/oh-my-opencode.schema.json
+++ b/assets/oh-my-opencode.schema.json
@@ -82,9 +82,6 @@
    "hashline_edit": {
      "type": "boolean"
    },
-    "model_fallback": {
-      "type": "boolean"
-    },
    "agents": {
      "type": "object",
      "properties": {
@@ -291,18 +288,6 @@
                }
              },
              "additionalProperties": false
-            },
-            "compaction": {
-              "type": "object",
-              "properties": {
-                "model": {
-                  "type": "string"
-                },
-                "variant": {
-                  "type": "string"
-                }
-              },
-              "additionalProperties": false
            }
          },
          "additionalProperties": false
@@ -510,18 +495,6 @@
                }
              },
              "additionalProperties": false
-            },
-            "compaction": {
-              "type": "object",
-              "properties": {
-                "model": {
-                  "type": "string"
-                },
-                "variant": {
-                  "type": "string"
-                }
-              },
-              "additionalProperties": false
            }
          },
          "additionalProperties": false
@@ -729,18 +702,6 @@
                }
              },
              "additionalProperties": false
-            },
-            "compaction": {
-              "type": "object",
-              "properties": {
-                "model": {
-                  "type": "string"
-                },
-                "variant": {
-                  "type": "string"
-                }
-              },
-              "additionalProperties": false
            }
          },
          "additionalProperties": false
@@ -948,21 +909,6 @@
                }
              },
              "additionalProperties": false
-            },
-            "compaction": {
-              "type": "object",
-              "properties": {
-                "model": {
-                  "type": "string"
-                },
-                "variant": {
-                  "type": "string"
-                }
-              },
-              "additionalProperties": false
-            },
-            "allow_non_gpt_model": {
-              "type": "boolean"
            }
          },
          "additionalProperties": false
@@ -1170,18 +1116,6 @@
                }
              },
              "additionalProperties": false
-            },
-            "compaction": {
-              "type": "object",
-              "properties": {
-                "model": {
-                  "type": "string"
-                },
-                "variant": {
-                  "type": "string"
-                }
-              },
-              "additionalProperties": false
            }
          },
          "additionalProperties": false
@@ -1389,18 +1323,6 @@
                }
              },
              "additionalProperties": false
-            },
-            "compaction": {
-              "type": "object",
-              "properties": {
-                "model": {
-                  "type": "string"
-                },
-                "variant": {
-                  "type": "string"
-                }
-              },
-              "additionalProperties": false
            }
          },
          "additionalProperties": false
@@ -1608,18 +1530,6 @@
                }
              },
              "additionalProperties": false
-            },
-            "compaction": {
-              "type": "object",
-              "properties": {
-                "model": {
-                  "type": "string"
-                },
-                "variant": {
-                  "type": "string"
-                }
-              },
-              "additionalProperties": false
            }
          },
          "additionalProperties": false
@@ -1827,18 +1737,6 @@
                }
              },
              "additionalProperties": false
-            },
-            "compaction": {
-              "type": "object",
-              "properties": {
-                "model": {
-                  "type": "string"
-                },
-                "variant": {
-                  "type": "string"
-                }
-              },
-              "additionalProperties": false
            }
          },
          "additionalProperties": false
@@ -2046,18 +1944,6 @@
                }
              },
              "additionalProperties": false
-            },
-            "compaction": {
-              "type": "object",
-              "properties": {
-                "model": {
-                  "type": "string"
-                },
-                "variant": {
-                  "type": "string"
-                }
-              },
-              "additionalProperties": false
            }
          },
          "additionalProperties": false
@@ -2265,18 +2151,6 @@
                }
              },
              "additionalProperties": false
-            },
-            "compaction": {
-              "type": "object",
-              "properties": {
-                "model": {
-                  "type": "string"
-                },
-                "variant": {
-                  "type": "string"
-                }
-              },
-              "additionalProperties": false
            }
          },
          "additionalProperties": false
@@ -2484,18 +2358,6 @@
                }
              },
              "additionalProperties": false
-            },
-            "compaction": {
-              "type": "object",
-              "properties": {
-                "model": {
-                  "type": "string"
-                },
-                "variant": {
-                  "type": "string"
-                }
-              },
-              "additionalProperties": false
            }
          },
          "additionalProperties": false
@@ -2703,18 +2565,6 @@
                }
              },
              "additionalProperties": false
-            },
-            "compaction": {
-              "type": "object",
-              "properties": {
-                "model": {
-                  "type": "string"
-                },
-                "variant": {
-                  "type": "string"
-                }
-              },
-              "additionalProperties": false
            }
          },
          "additionalProperties": false
@@ -2922,18 +2772,6 @@
                }
              },
              "additionalProperties": false
-            },
-            "compaction": {
-              "type": "object",
-              "properties": {
-                "model": {
-                  "type": "string"
-                },
-                "variant": {
-                  "type": "string"
-                }
-              },
-              "additionalProperties": false
            }
          },
          "additionalProperties": false
@@ -3141,18 +2979,6 @@
                }
              },
              "additionalProperties": false
-            },
-            "compaction": {
-              "type": "object",
-              "properties": {
-                "model": {
-                  "type": "string"
-                },
-                "variant": {
-                  "type": "string"
-                }
-              },
-              "additionalProperties": false
            }
          },
          "additionalProperties": false
@@ -3251,11 +3077,6 @@
          "prompt_append": {
            "type": "string"
          },
-          "max_prompt_tokens": {
-            "type": "integer",
-            "exclusiveMinimum": 0,
-            "maximum": 9007199254740991
-          },
          "is_unstable_agent": {
            "type": "boolean"
          },
--- a/bun.lock
+++ b/bun.lock
@@ -14,7 +14,6 @@
        "@opencode-ai/sdk": "^1.1.19",
        "commander": "^14.0.2",
        "detect-libc": "^2.0.0",
-        "diff": "^8.0.3",
        "js-yaml": "^4.1.1",
        "jsonc-parser": "^3.3.1",
        "picocolors": "^1.1.1",
@@ -29,13 +28,13 @@
        "typescript": "^5.7.3",
      },
      "optionalDependencies": {
-        "oh-my-opencode-darwin-arm64": "3.8.5",
-        "oh-my-opencode-darwin-x64": "3.8.5",
-        "oh-my-opencode-linux-arm64": "3.8.5",
-        "oh-my-opencode-linux-arm64-musl": "3.8.5",
-        "oh-my-opencode-linux-x64": "3.8.5",
-        "oh-my-opencode-linux-x64-musl": "3.8.5",
-        "oh-my-opencode-windows-x64": "3.8.5",
+        "oh-my-opencode-darwin-arm64": "3.7.4",
+        "oh-my-opencode-darwin-x64": "3.7.4",
+        "oh-my-opencode-linux-arm64": "3.7.4",
+        "oh-my-opencode-linux-arm64-musl": "3.7.4",
+        "oh-my-opencode-linux-x64": "3.7.4",
+        "oh-my-opencode-linux-x64-musl": "3.7.4",
+        "oh-my-opencode-windows-x64": "3.7.4",
      },
    },
  },
@@ -139,8 +138,6 @@

    "detect-libc": ["detect-libc@2.1.2", "", {}, "sha512-Btj2BOOO83o3WyH59e8MgXsxEQVcarkUOpEYrubB0urwnN10yQ364rsiByU11nZlqWYZm05i/of7io4mzihBtQ=="],

-    "diff": ["diff@8.0.3", "", {}, "sha512-qejHi7bcSD4hQAZE0tNAawRK1ZtafHDmMTMkrrIGgSLl7hTnQHmKCeB45xAcbfTqK2zowkM3j3bHt/4b/ARbYQ=="],
-
    "dunder-proto": ["dunder-proto@1.0.1", "", { "dependencies": { "call-bind-apply-helpers": "^1.0.1", "es-errors": "^1.3.0", "gopd": "^1.2.0" } }, "sha512-KIN/nDJBQRcXw0MLVhZE9iQHmG68qAVIBg9CqmUYjmQIhgij9U5MFvrqkUL5FbtyyzZuOeOt0zdeRe4UY7ct+A=="],

    "ee-first": ["ee-first@1.1.1", "", {}, "sha512-WMwm9LhRUo+WUaRN+vRuETqG89IgZphVSNkdFgeb6sS/E4OrDIN7t48CAewSHXc6C8lefD8KKfr5vY61brQlow=="],
@@ -231,19 +228,19 @@

    "object-inspect": ["object-inspect@1.13.4", "", {}, "sha512-W67iLl4J2EXEGTbfeHCffrjDfitvLANg0UlX3wFUUSTx92KXRFegMHUVgSqE+wvhAbi4WqjGg9czysTV2Epbew=="],

-    "oh-my-opencode-darwin-arm64": ["oh-my-opencode-darwin-arm64@3.8.5", "", { "os": "darwin", "cpu": "arm64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-bbLu1We9NNhYAVp9Q/FK8dYFlYLp2PKfvdBCr+O6QjNRixdjp8Ru4RK7i9mKg0ybYBUzzCcbbC2Cc1o8orkhBA=="],
+    "oh-my-opencode-darwin-arm64": ["oh-my-opencode-darwin-arm64@3.7.4", "", { "os": "darwin", "cpu": "arm64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-0m84UiVlOC2gLSFIOTmCsxFCB9CmyWV9vGPYqfBFLoyDJmedevU3R5N4ze54W7jv4HSSxz02Zwr+QF5rkQANoA=="],

-    "oh-my-opencode-darwin-x64": ["oh-my-opencode-darwin-x64@3.8.5", "", { "os": "darwin", "cpu": "x64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-N9GcmzYgL87UybSaMGiHc5lwT5Mxg1tyB502el5syouN39wfeUYoj37SonENrMUTiEfn75Lwv/5cSLCesSubpA=="],
+    "oh-my-opencode-darwin-x64": ["oh-my-opencode-darwin-x64@3.7.4", "", { "os": "darwin", "cpu": "x64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-Z2dQy8jmc6DuwbN9bafhOwjZBkAkTWlfLAz1tG6xVzMqTcp4YOrzrHFOBRNeFKpOC/x7yUpO3sq/YNCclloelw=="],

-    "oh-my-opencode-linux-arm64": ["oh-my-opencode-linux-arm64@3.8.5", "", { "os": "linux", "cpu": "arm64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-ki4a7s1DD5z5wEKmzcchqAKOIpw0LsBvyF8ieqNLS5Xl8PWE0gAZ7rqjlXC54NTubpexVH6lO2yenFJsk2Zk9A=="],
+    "oh-my-opencode-linux-arm64": ["oh-my-opencode-linux-arm64@3.7.4", "", { "os": "linux", "cpu": "arm64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-TZIsK6Dl6yX6pSTocls91bjnvoY/6/kiGnmgdsoDKcPYZ7XuBQaJwH0dK7t9/sxuDI+wKhmtrmLwKSoYOIqsRw=="],

-    "oh-my-opencode-linux-arm64-musl": ["oh-my-opencode-linux-arm64-musl@3.8.5", "", { "os": "linux", "cpu": "arm64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-9+6hU3z503fBzuV0VjxIkTKFElbKacHijFcdKAussG6gPFLWmCRWtdowzEDwUfAoIsoHHH7FBwvh5waGp/ZksA=="],
+    "oh-my-opencode-linux-arm64-musl": ["oh-my-opencode-linux-arm64-musl@3.7.4", "", { "os": "linux", "cpu": "arm64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-UwPOoQP0+1eCKP/XTDsnLJDK5jayiL4VrKz0lfRRRojl1FWvInmQumnDnluvnxW6knU7dFM3yDddlZYG6tEgcw=="],

-    "oh-my-opencode-linux-x64": ["oh-my-opencode-linux-x64@3.8.5", "", { "os": "linux", "cpu": "x64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-DmnMK/PgvdcCYL+OQE5iZWgi/vmjm0sIPQVQgSUbWn3izcUF7C5DtlxqaU2cKxNZwrhDTlJdLWxmJqgLmLqd9A=="],
+    "oh-my-opencode-linux-x64": ["oh-my-opencode-linux-x64@3.7.4", "", { "os": "linux", "cpu": "x64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-+TeA0Bs5wK9EMfKiEEFfyfVqdBDUjDzN8POF8JJibN0GPy1oNIGGEWIJG2cvC5onpnYEvl448vkFbkCUK0g9SQ=="],

-    "oh-my-opencode-linux-x64-musl": ["oh-my-opencode-linux-x64-musl@3.8.5", "", { "os": "linux", "cpu": "x64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-jhCNStljsyapVq9X7PaHSOcWxxEA4BUcIibvoPs/xc7fVP8D47p651LzIRsM6STn6Bx684mlYbxxX1P/0QPKNg=="],
+    "oh-my-opencode-linux-x64-musl": ["oh-my-opencode-linux-x64-musl@3.7.4", "", { "os": "linux", "cpu": "x64", "bin": { "oh-my-opencode": "bin/oh-my-opencode" } }, "sha512-YzX6wFtk8RoTHkAZkfLCVyCU4yjN8D7agj/jhOnFKW50fZYa8zX+/4KLZx0IfanVpXTgrs3iiuKoa87KLDfCxQ=="],

-    "oh-my-opencode-windows-x64": ["oh-my-opencode-windows-x64@3.8.5", "", { "os": "win32", "cpu": "x64", "bin": { "oh-my-opencode": "bin/oh-my-opencode.exe" } }, "sha512-lcPBp9NCNQ6TnqzsN9p/K+xKwOzBoIPw7HncxmrXSberZ3uHy0K9uNraQ7fqnXIKWqQiK4kSwWfSHpmhbaHiNg=="],
+    "oh-my-opencode-windows-x64": ["oh-my-opencode-windows-x64@3.7.4", "", { "os": "win32", "cpu": "x64", "bin": { "oh-my-opencode": "bin/oh-my-opencode.exe" } }, "sha512-x39M2eFJI6pqv4go5Crf1H2SbPGFmXHIDNtbsSa5nRNcrqTisLrYGW8uXpOrqjntBeTAUBdwZmmoy6zgxHsz8w=="],

    "on-finished": ["on-finished@2.4.1", "", { "dependencies": { "ee-first": "1.1.1" } }, "sha512-oVlzkg3ENAhCk2zdv7IJwd/QUD4z2RxRwpkcGY8psCVcCYZNq4wYnVWALHM+brtuJjePWiYF/ClmuDr8Ch5+kg=="],

--- a/package.json
+++ b/package.json
@@ -1,6 +1,6 @@
 {
  "name": "oh-my-opencode",
-  "version": "3.8.5",
+  "version": "3.8.1",
  "description": "The Best AI Agent Harness - Batteries-Included OpenCode Plugin with Multi-Model Orchestration, Parallel Background Agents, and Crafted LSP/AST Tools",
  "main": "dist/index.js",
  "types": "dist/index.d.ts",
@@ -60,7 +60,6 @@
    "@opencode-ai/sdk": "^1.1.19",
    "commander": "^14.0.2",
    "detect-libc": "^2.0.0",
-    "diff": "^8.0.3",
    "js-yaml": "^4.1.1",
    "jsonc-parser": "^3.3.1",
    "picocolors": "^1.1.1",
@@ -75,13 +74,13 @@
    "typescript": "^5.7.3"
  },
  "optionalDependencies": {
-    "oh-my-opencode-darwin-arm64": "3.8.5",
-    "oh-my-opencode-darwin-x64": "3.8.5",
-    "oh-my-opencode-linux-arm64": "3.8.5",
-    "oh-my-opencode-linux-arm64-musl": "3.8.5",
-    "oh-my-opencode-linux-x64": "3.8.5",
-    "oh-my-opencode-linux-x64-musl": "3.8.5",
-    "oh-my-opencode-windows-x64": "3.8.5"
+    "oh-my-opencode-darwin-arm64": "3.8.1",
+    "oh-my-opencode-darwin-x64": "3.8.1",
+    "oh-my-opencode-linux-arm64": "3.8.1",
+    "oh-my-opencode-linux-arm64-musl": "3.8.1",
+    "oh-my-opencode-linux-x64": "3.8.1",
+    "oh-my-opencode-linux-x64-musl": "3.8.1",
+    "oh-my-opencode-windows-x64": "3.8.1"
  },
  "trustedDependencies": [
    "@ast-grep/cli",
--- a/packages/darwin-arm64/package.json
+++ b/packages/darwin-arm64/package.json
@@ -1,6 +1,6 @@
 {
  "name": "oh-my-opencode-darwin-arm64",
-  "version": "3.8.5",
+  "version": "3.8.1",
  "description": "Platform-specific binary for oh-my-opencode (darwin-arm64)",
  "license": "MIT",
  "repository": {
--- a/packages/darwin-x64/package.json
+++ b/packages/darwin-x64/package.json
@@ -1,6 +1,6 @@
 {
  "name": "oh-my-opencode-darwin-x64",
-  "version": "3.8.5",
+  "version": "3.8.1",
  "description": "Platform-specific binary for oh-my-opencode (darwin-x64)",
  "license": "MIT",
  "repository": {
--- a/packages/linux-arm64-musl/package.json
+++ b/packages/linux-arm64-musl/package.json
@@ -1,6 +1,6 @@
 {
  "name": "oh-my-opencode-linux-arm64-musl",
-  "version": "3.8.5",
+  "version": "3.8.1",
  "description": "Platform-specific binary for oh-my-opencode (linux-arm64-musl)",
  "license": "MIT",
  "repository": {
--- a/packages/linux-arm64/package.json
+++ b/packages/linux-arm64/package.json
@@ -1,6 +1,6 @@
 {
  "name": "oh-my-opencode-linux-arm64",
-  "version": "3.8.5",
+  "version": "3.8.1",
  "description": "Platform-specific binary for oh-my-opencode (linux-arm64)",
  "license": "MIT",
  "repository": {
--- a/packages/linux-x64-musl/package.json
+++ b/packages/linux-x64-musl/package.json
@@ -1,6 +1,6 @@
 {
  "name": "oh-my-opencode-linux-x64-musl",
-  "version": "3.8.5",
+  "version": "3.8.1",
  "description": "Platform-specific binary for oh-my-opencode (linux-x64-musl)",
  "license": "MIT",
  "repository": {
--- a/packages/linux-x64/package.json
+++ b/packages/linux-x64/package.json
@@ -1,6 +1,6 @@
 {
  "name": "oh-my-opencode-linux-x64",
-  "version": "3.8.5",
+  "version": "3.8.1",
  "description": "Platform-specific binary for oh-my-opencode (linux-x64)",
  "license": "MIT",
  "repository": {
--- a/packages/windows-x64/package.json
+++ b/packages/windows-x64/package.json
@@ -1,6 +1,6 @@
 {
  "name": "oh-my-opencode-windows-x64",
-  "version": "3.8.5",
+  "version": "3.8.1",
  "description": "Platform-specific binary for oh-my-opencode (windows-x64)",
  "license": "MIT",
  "repository": {
--- a/signatures/cla.json
+++ b/signatures/cla.json
@@ -1679,94 +1679,6 @@
      "created_at": "2026-02-21T22:44:45Z",
      "repoId": 1108837393,
      "pullRequestNo": 2029
-    },
-    {
-      "name": "imadal1n",
-      "id": 97968636,
-      "comment_id": 3940704780,
-      "created_at": "2026-02-22T10:57:33Z",
-      "repoId": 1108837393,
-      "pullRequestNo": 2045
-    },
-    {
-      "name": "DMax1314",
-      "id": 54206290,
-      "comment_id": 3943046087,
-      "created_at": "2026-02-23T07:06:14Z",
-      "repoId": 1108837393,
-      "pullRequestNo": 2068
-    },
-    {
-      "name": "Firstbober",
-      "id": 22197465,
-      "comment_id": 3946848526,
-      "created_at": "2026-02-23T19:27:59Z",
-      "repoId": 1108837393,
-      "pullRequestNo": 2080
-    },
-    {
-      "name": "PHP-Expert",
-      "id": 12047666,
-      "comment_id": 3951828700,
-      "created_at": "2026-02-24T13:27:18Z",
-      "repoId": 1108837393,
-      "pullRequestNo": 2098
-    },
-    {
-      "name": "Pantoria",
-      "id": 37699442,
-      "comment_id": 3953543578,
-      "created_at": "2026-02-24T17:12:31Z",
-      "repoId": 1108837393,
-      "pullRequestNo": 1983
-    },
-    {
-      "name": "east-shine",
-      "id": 20237288,
-      "comment_id": 3957576758,
-      "created_at": "2026-02-25T08:19:34Z",
-      "repoId": 1108837393,
-      "pullRequestNo": 2113
-    },
-    {
-      "name": "SupenBysz",
-      "id": 3314033,
-      "comment_id": 3962352704,
-      "created_at": "2026-02-25T22:00:54Z",
-      "repoId": 1108837393,
-      "pullRequestNo": 2119
-    },
-    {
-      "name": "zhzy0077",
-      "id": 8717471,
-      "comment_id": 3964015975,
-      "created_at": "2026-02-26T04:45:23Z",
-      "repoId": 1108837393,
-      "pullRequestNo": 2125
-    },
-    {
-      "name": "spacecowboy0416",
-      "id": 239068998,
-      "comment_id": 3964320737,
-      "created_at": "2026-02-26T06:05:27Z",
-      "repoId": 1108837393,
-      "pullRequestNo": 2126
-    },
-    {
-      "name": "imwxc",
-      "id": 49653609,
-      "comment_id": 3965127447,
-      "created_at": "2026-02-26T09:00:16Z",
-      "repoId": 1108837393,
-      "pullRequestNo": 2129
-    },
-    {
-      "name": "maou-shonen",
-      "id": 22576780,
-      "comment_id": 3965445132,
-      "created_at": "2026-02-26T09:50:46Z",
-      "repoId": 1108837393,
-      "pullRequestNo": 2131
    }
  ]
 }
--- a/src/AGENTS.md
+++ b/src/AGENTS.md
@@ -1,6 +1,6 @@
 # src/ — Plugin Source

-**Generated:** 2026-02-24
+**Generated:** 2026-02-21

 ## OVERVIEW

@@ -14,7 +14,7 @@ Root source directory. Entry point `index.ts` orchestrates 4-step initialization
 | `plugin-config.ts` | JSONC parse, multi-level merge (user → project → defaults), Zod validation |
 | `create-managers.ts` | TmuxSessionManager, BackgroundManager, SkillMcpManager, ConfigHandler |
 | `create-tools.ts` | SkillContext + AvailableCategories + ToolRegistry |
-| `create-hooks.ts` | 3-tier hook composition: Core(37) + Continuation(7) + Skill(2) |
+| `create-hooks.ts` | 3-tier hook composition: Core(35) + Continuation(7) + Skill(2) |
 | `plugin-interface.ts` | Assembles 8 OpenCode hook handlers into PluginInterface |

 ## CONFIG LOADING
@@ -32,9 +32,9 @@ loadPluginConfig(directory, ctx)

 ```
 createHooks()
-  ├─→ createCoreHooks()           # 37 hooks
-  │   ├─ createSessionHooks()     # 23: contextWindowMonitor, thinkMode, ralphLoop, modelFallback, runtimeFallback, noSisyphusGpt, noHephaestusNonGpt, anthropicEffort...
-  │   ├─ createToolGuardHooks()   # 10: commentChecker, rulesInjector, writeExistingFileGuard, jsonErrorRecovery, hashlineReadEnhancer...
+  ├─→ createCoreHooks()           # 35 hooks
+  │   ├─ createSessionHooks()     # 21: contextWindowMonitor, thinkMode, ralphLoop, sessionRecovery, jsonErrorRecovery, sisyphusGptHephaestusReminder, anthropicEffort...
+  │   ├─ createToolGuardHooks()   # 10: commentChecker, rulesInjector, writeExistingFileGuard, hashlineEditDiffEnhancer...
  │   └─ createTransformHooks()   # 4: claudeCodeHooks, keywordDetector, contextInjector, thinkingBlockValidator
  ├─→ createContinuationHooks()   # 7: todoContinuationEnforcer, atlas, stopContinuationGuard...
  └─→ createSkillHooks()          # 2: categorySkillReminder, autoSlashCommand
--- a/src/agents/AGENTS.md
+++ b/src/agents/AGENTS.md
@@ -1,6 +1,6 @@
 # src/agents/ — 11 Agent Definitions

-**Generated:** 2026-02-24
+**Generated:** 2026-02-21

 ## OVERVIEW

--- a/src/agents/atlas/agent.ts
+++ b/src/agents/atlas/agent.ts
@@ -6,21 +6,20 @@
 *
 * Routing:
 * 1. GPT models (openai/*, github-copilot/gpt-*) → gpt.ts (GPT-5.2 optimized)
- * 2. Gemini models (google/*, google-vertex/*) → gemini.ts (Gemini-optimized)
- * 3. Default (Claude, etc.) → default.ts (Claude-optimized)
+ * 2. Default (Claude, etc.) → default.ts (Claude-optimized)
 */

 import type { AgentConfig } from "@opencode-ai/sdk"
 import type { AgentMode, AgentPromptMetadata } from "../types"
-import { isGptModel, isGeminiModel } from "../types"
+import { isGptModel } from "../types"
 import type { AvailableAgent, AvailableSkill, AvailableCategory } from "../dynamic-agent-prompt-builder"
 import { buildCategorySkillsDelegationGuide } from "../dynamic-agent-prompt-builder"
 import type { CategoryConfig } from "../../config/schema"
 import { mergeCategories } from "../../shared/merge-categories"
+import { createAgentToolRestrictions } from "../../shared/permission-compat"

 import { getDefaultAtlasPrompt } from "./default"
 import { getGptAtlasPrompt } from "./gpt"
-import { getGeminiAtlasPrompt } from "./gemini"
 import {
  getCategoryDescription,
  buildAgentSelectionSection,
@@ -31,7 +30,7 @@ import {

 const MODE: AgentMode = "primary"

-export type AtlasPromptSource = "default" | "gpt" | "gemini"
+export type AtlasPromptSource = "default" | "gpt"

 /**
 * Determines which Atlas prompt to use based on model.
@@ -40,9 +39,6 @@ export function getAtlasPromptSource(model?: string): AtlasPromptSource {
  if (model && isGptModel(model)) {
    return "gpt"
  }
-  if (model && isGeminiModel(model)) {
-    return "gemini"
-  }
  return "default"
 }

@@ -62,8 +58,6 @@ export function getAtlasPrompt(model?: string): string {
  switch (source) {
    case "gpt":
      return getGptAtlasPrompt()
-    case "gemini":
-      return getGeminiAtlasPrompt()
    case "default":
    default:
      return getDefaultAtlasPrompt()
@@ -99,6 +93,11 @@ function buildDynamicOrchestratorPrompt(ctx?: OrchestratorContext): string {
 }

 export function createAtlasAgent(ctx: OrchestratorContext): AgentConfig {
+  const restrictions = createAgentToolRestrictions([
+    "task",
+    "call_omo_agent",
+  ])
+
  const baseConfig = {
    description:
      "Orchestrates work via task() to complete ALL tasks in a todo list until fully done. (Atlas - OhMyOpenCode)",
@@ -107,6 +106,7 @@ export function createAtlasAgent(ctx: OrchestratorContext): AgentConfig {
    temperature: 0.1,
    prompt: buildDynamicOrchestratorPrompt(ctx),
    color: "#10B981",
+    ...restrictions,
  }

  return baseConfig as AgentConfig
--- a/src/agents/atlas/index.ts
+++ b/src/agents/atlas/index.ts
@@ -1,2 +1,14 @@
-export { createAtlasAgent, atlasPromptMetadata } from "./agent"
+export { ATLAS_SYSTEM_PROMPT, getDefaultAtlasPrompt } from "./default"
+export { ATLAS_GPT_SYSTEM_PROMPT, getGptAtlasPrompt } from "./gpt"
+export {
+  getCategoryDescription,
+  buildAgentSelectionSection,
+  buildCategorySection,
+  buildSkillsSection,
+  buildDecisionMatrix,
+} from "./prompt-section-builder"
+
+export { createAtlasAgent, getAtlasPromptSource, getAtlasPrompt, atlasPromptMetadata } from "./agent"
 export type { AtlasPromptSource, OrchestratorContext } from "./agent"
+
+export { isGptModel } from "../types"
--- a/src/agents/env-context.test.ts
+++ b/src/agents/env-context.test.ts
@@ -1,41 +0,0 @@
-/// <reference types="bun-types" />
-
-import { describe, test, expect } from "bun:test"
-import { createEnvContext } from "./env-context"
-
-describe("createEnvContext", () => {
-  test("returns omo-env block with timezone and locale", () => {
-    // #given - no setup needed
-
-    // #when
-    const result = createEnvContext()
-
-    // #then
-    expect(result).toContain("<omo-env>")
-    expect(result).toContain("</omo-env>")
-    expect(result).toContain("Timezone:")
-    expect(result).toContain("Locale:")
-    expect(result).not.toContain("Current date:")
-  })
-
-  test("does not include time with seconds precision to preserve token cache", () => {
-    // #given - seconds-precision time changes every second, breaking cache on every request
-
-    // #when
-    const result = createEnvContext()
-
-    // #then - no HH:MM:SS pattern anywhere in the output
-    expect(result).not.toMatch(/\d{1,2}:\d{2}:\d{2}/)
-  })
-
-  test("does not include date or time fields since OpenCode already provides them", () => {
-    // #given - OpenCode's system.ts already injects date, platform, working directory
-
-    // #when
-    const result = createEnvContext()
-
-    // #then - only timezone and locale remain; both are stable across requests
-    expect(result).not.toContain("Current date:")
-    expect(result).not.toContain("Current time:")
-  })
-})
--- a/src/agents/env-context.ts
+++ b/src/agents/env-context.ts
@@ -1,15 +1,32 @@
 /**
- * Creates OmO-specific environment context (timezone, locale).
+ * Creates OmO-specific environment context (time, timezone, locale).
 * Note: Working directory, platform, and date are already provided by OpenCode's system.ts,
 * so we only include fields that OpenCode doesn't provide to avoid duplication.
 * See: https://github.com/code-yeongyu/oh-my-opencode/issues/379
 */
 export function createEnvContext(): string {
+  const now = new Date()
  const timezone = Intl.DateTimeFormat().resolvedOptions().timeZone
  const locale = Intl.DateTimeFormat().resolvedOptions().locale

+  const dateStr = now.toLocaleDateString(locale, {
+    weekday: "short",
+    year: "numeric",
+    month: "short",
+    day: "numeric",
+  })
+
+  const timeStr = now.toLocaleTimeString(locale, {
+    hour: "2-digit",
+    minute: "2-digit",
+    second: "2-digit",
+    hour12: true,
+  })
+
  return `
 <omo-env>
+  Current date: ${dateStr}
+  Current time: ${timeStr}
  Timezone: ${timezone}
  Locale: ${locale}
 </omo-env>`
--- a/src/agents/hephaestus.ts
+++ b/src/agents/hephaestus.ts
@@ -448,21 +448,6 @@ ${oracleSection}
 4. **Run build** if applicable — exit code 0 required
 5. **Tell user** what you verified and the results — keep it clear and helpful

-### Auto-Commit Policy (MANDATORY for implementation/fix work)
-
-1. **Auto-commit after implementation is complete** when the task includes feature/fix code changes
-2. **Commit ONLY after verification gates pass**:
-   - \`lsp_diagnostics\` clean on all modified files
-   - Related tests pass
-   - Typecheck/build pass when applicable
-3. **If any gate fails, DO NOT commit** — fix issues first, re-run verification, then commit
-4. **Use Conventional Commits format** with meaningful intent-focused messages:
-   - \`feat(scope): add ...\` for new functionality
-   - \`fix(scope): resolve ...\` for bug fixes
-   - \`refactor(scope): simplify ...\` for internal restructuring
-5. **Do not make placeholder commits** (\`wip\`, \`temp\`, \`update\`) or commit unverified code
-6. **If user explicitly says not to commit**, skip commit and report that changes are left uncommitted
-
 - **File edit** — \`lsp_diagnostics\` clean
 - **Build** — Exit code 0
 - **Tests** — Pass (or pre-existing failures noted)
--- a/src/agents/index.ts
+++ b/src/agents/index.ts
@@ -1,4 +1,28 @@
 export * from "./types"
 export { createBuiltinAgents } from "./builtin-agents"
 export type { AvailableAgent, AvailableCategory, AvailableSkill } from "./dynamic-agent-prompt-builder"
+export { createSisyphusAgent } from "./sisyphus"
+export { createOracleAgent, ORACLE_PROMPT_METADATA } from "./oracle"
+export { createLibrarianAgent, LIBRARIAN_PROMPT_METADATA } from "./librarian"
+export { createExploreAgent, EXPLORE_PROMPT_METADATA } from "./explore"
+
+
+export { createMultimodalLookerAgent, MULTIMODAL_LOOKER_PROMPT_METADATA } from "./multimodal-looker"
+export { createMetisAgent, METIS_SYSTEM_PROMPT, metisPromptMetadata } from "./metis"
+export { createMomusAgent, MOMUS_SYSTEM_PROMPT, momusPromptMetadata } from "./momus"
+export { createAtlasAgent, atlasPromptMetadata } from "./atlas"
+export {
+  PROMETHEUS_SYSTEM_PROMPT,
+  PROMETHEUS_PERMISSION,
+  PROMETHEUS_GPT_SYSTEM_PROMPT,
+  getPrometheusPrompt,
+  getPrometheusPromptSource,
+  getGptPrometheusPrompt,
+  PROMETHEUS_IDENTITY_CONSTRAINTS,
+  PROMETHEUS_INTERVIEW_MODE,
+  PROMETHEUS_PLAN_GENERATION,
+  PROMETHEUS_HIGH_ACCURACY_MODE,
+  PROMETHEUS_PLAN_TEMPLATE,
+  PROMETHEUS_BEHAVIORAL_SUMMARY,
+} from "./prometheus"
 export type { PrometheusPromptSource } from "./prometheus"
--- a/src/agents/prometheus/index.ts
+++ b/src/agents/prometheus/index.ts
@@ -2,5 +2,15 @@ export {
  PROMETHEUS_SYSTEM_PROMPT,
  PROMETHEUS_PERMISSION,
  getPrometheusPrompt,
+  getPrometheusPromptSource,
 } from "./system-prompt"
 export type { PrometheusPromptSource } from "./system-prompt"
+export { PROMETHEUS_GPT_SYSTEM_PROMPT, getGptPrometheusPrompt } from "./gpt"
+
+// Re-export individual sections for granular access
+export { PROMETHEUS_IDENTITY_CONSTRAINTS } from "./identity-constraints"
+export { PROMETHEUS_INTERVIEW_MODE } from "./interview-mode"
+export { PROMETHEUS_PLAN_GENERATION } from "./plan-generation"
+export { PROMETHEUS_HIGH_ACCURACY_MODE } from "./high-accuracy-mode"
+export { PROMETHEUS_PLAN_TEMPLATE } from "./plan-template"
+export { PROMETHEUS_BEHAVIORAL_SUMMARY } from "./behavioral-summary"
--- a/src/agents/prometheus/system-prompt.ts
+++ b/src/agents/prometheus/system-prompt.ts
@@ -5,8 +5,7 @@ import { PROMETHEUS_HIGH_ACCURACY_MODE } from "./high-accuracy-mode"
 import { PROMETHEUS_PLAN_TEMPLATE } from "./plan-template"
 import { PROMETHEUS_BEHAVIORAL_SUMMARY } from "./behavioral-summary"
 import { getGptPrometheusPrompt } from "./gpt"
-import { getGeminiPrometheusPrompt } from "./gemini"
-import { isGptModel, isGeminiModel } from "../types"
+import { isGptModel } from "../types"

 /**
 * Combined Prometheus system prompt (Claude-optimized, default).
@@ -31,7 +30,7 @@ export const PROMETHEUS_PERMISSION = {
  question: "allow" as const,
 }

-export type PrometheusPromptSource = "default" | "gpt" | "gemini"
+export type PrometheusPromptSource = "default" | "gpt"

 /**
 * Determines which Prometheus prompt to use based on model.
@@ -40,16 +39,12 @@ export function getPrometheusPromptSource(model?: string): PrometheusPromptSourc
  if (model && isGptModel(model)) {
    return "gpt"
  }
-  if (model && isGeminiModel(model)) {
-    return "gemini"
-  }
  return "default"
 }

 /**
 * Gets the appropriate Prometheus prompt based on model.
 * GPT models → GPT-5.2 optimized prompt (XML-tagged, principle-driven)
- * Gemini models → Gemini-optimized prompt (aggressive tool-call enforcement, thinking checkpoints)
 * Default (Claude, etc.) → Claude-optimized prompt (modular sections)
 */
 export function getPrometheusPrompt(model?: string): string {
@@ -58,8 +53,6 @@ export function getPrometheusPrompt(model?: string): string {
  switch (source) {
    case "gpt":
      return getGptPrometheusPrompt()
-    case "gemini":
-      return getGeminiPrometheusPrompt()
    case "default":
    default:
      return PROMETHEUS_SYSTEM_PROMPT
--- a/src/agents/sisyphus-gemini-overlays.ts
+++ b/src/agents/sisyphus-gemini-overlays.ts
@@ -6,8 +6,6 @@
 * - Avoid delegation, preferring to do work themselves
 * - Claim completion without verification
 * - Interpret constraints as suggestions
- * - Skip intent classification gates (jump straight to action)
- * - Conflate investigation with implementation ("look into X" → starts coding)
 *
 * These overlays inject corrective sections at strategic points
 * in the dynamic Sisyphus prompt to counter these tendencies.
@@ -79,39 +77,3 @@ Your internal confidence estimator is miscalibrated toward optimism. What feels
 4. If you delegated, read EVERY file the subagent touched — not trust their claims
 </GEMINI_VERIFICATION_OVERRIDE>`;
 }
-
-export function buildGeminiIntentGateEnforcement(): string {
-  return `<GEMINI_INTENT_GATE_ENFORCEMENT>
-## YOU MUST CLASSIFY INTENT BEFORE ACTING. NO EXCEPTIONS.
-
-**Your failure mode: You skip intent classification and jump straight to implementation.**
-
-You see a user message and your instinct is to immediately start working. WRONG. You MUST first determine WHAT KIND of work the user wants. Getting this wrong wastes everything that follows.
-
-**MANDATORY FIRST OUTPUT — before ANY tool call or action:**
-
-\`\`\`
-I detect [TYPE] intent — [REASON].
-My approach: [ROUTING DECISION].
-\`\`\`
-
-Where TYPE is one of: research | implementation | investigation | evaluation | fix | open-ended
-
-**SELF-CHECK (answer honestly before proceeding):**
-
-1. Did the user EXPLICITLY ask me to implement/build/create something? → If NO, do NOT implement.
-2. Did the user say "look into", "check", "investigate", "explain"? → That means RESEARCH, not implementation.
-3. Did the user ask "what do you think?" → That means EVALUATION — propose and WAIT, do not execute.
-4. Did the user report an error? → That means MINIMAL FIX, not refactoring.
-
-**COMMON MISTAKES YOU MAKE (AND MUST NOT):**
-
-| User Says | You Want To Do | You MUST Do |
-| "explain how X works" | Start modifying X | Research X, explain it, STOP |
-| "look into this bug" | Fix the bug immediately | Investigate, report findings, WAIT for go-ahead |
-| "what do you think about approach X?" | Implement approach X | Evaluate X, propose alternatives, WAIT |
-| "improve the tests" | Rewrite all tests | Assess current tests FIRST, propose approach, THEN implement |
-
-**IF YOU SKIPPED THE INTENT CLASSIFICATION ABOVE:** STOP. Go back. Do it now. Your next tool call is INVALID without it.
-</GEMINI_INTENT_GATE_ENFORCEMENT>`;
-}
--- a/src/agents/sisyphus-junior/agent.ts
+++ b/src/agents/sisyphus-junior/agent.ts
@@ -6,13 +6,12 @@
 *
 * Routing:
 * 1. GPT models (openai/*, github-copilot/gpt-*) -> gpt.ts (GPT-5.2 optimized)
- * 2. Gemini models (google/*, google-vertex/*) -> gemini.ts (Gemini-optimized)
- * 3. Default (Claude, etc.) -> default.ts (Claude-optimized)
+ * 2. Default (Claude, etc.) -> default.ts (Claude-optimized)
 */

 import type { AgentConfig } from "@opencode-ai/sdk"
 import type { AgentMode } from "../types"
-import { isGptModel, isGeminiModel } from "../types"
+import { isGptModel } from "../types"
 import type { AgentOverrideConfig } from "../../config/schema"
 import {
  createAgentToolRestrictions,
@@ -21,7 +20,6 @@ import {

 import { buildDefaultSisyphusJuniorPrompt } from "./default"
 import { buildGptSisyphusJuniorPrompt } from "./gpt"
-import { buildGeminiSisyphusJuniorPrompt } from "./gemini"

 const MODE: AgentMode = "subagent"

@@ -34,7 +32,7 @@ export const SISYPHUS_JUNIOR_DEFAULTS = {
  temperature: 0.1,
 } as const

-export type SisyphusJuniorPromptSource = "default" | "gpt" | "gemini"
+export type SisyphusJuniorPromptSource = "default" | "gpt"

 /**
 * Determines which Sisyphus-Junior prompt to use based on model.
@@ -43,9 +41,6 @@ export function getSisyphusJuniorPromptSource(model?: string): SisyphusJuniorPro
  if (model && isGptModel(model)) {
    return "gpt"
  }
-  if (model && isGeminiModel(model)) {
-    return "gemini"
-  }
  return "default"
 }

@@ -62,8 +57,6 @@ export function buildSisyphusJuniorPrompt(
  switch (source) {
    case "gpt":
      return buildGptSisyphusJuniorPrompt(useTaskSystem, promptAppend)
-    case "gemini":
-      return buildGeminiSisyphusJuniorPrompt(useTaskSystem, promptAppend)
    case "default":
    default:
      return buildDefaultSisyphusJuniorPrompt(useTaskSystem, promptAppend)
--- a/src/agents/sisyphus-junior/index.ts
+++ b/src/agents/sisyphus-junior/index.ts
@@ -1,6 +1,5 @@
 export { buildDefaultSisyphusJuniorPrompt } from "./default"
 export { buildGptSisyphusJuniorPrompt } from "./gpt"
-export { buildGeminiSisyphusJuniorPrompt } from "./gemini"

 export {
  SISYPHUS_JUNIOR_DEFAULTS,
--- a/src/agents/sisyphus.ts
+++ b/src/agents/sisyphus.ts
@@ -1,12 +1,6 @@
 import type { AgentConfig } from "@opencode-ai/sdk";
 import type { AgentMode, AgentPromptMetadata } from "./types";
-import { isGptModel, isGeminiModel } from "./types";
-import {
-  buildGeminiToolMandate,
-  buildGeminiDelegationOverride,
-  buildGeminiVerificationOverride,
-  buildGeminiIntentGateEnforcement,
-} from "./sisyphus-gemini-overlays";
+import { isGptModel } from "./types";

 const MODE: AgentMode = "primary";
 export const SISYPHUS_PROMPT_METADATA: AgentPromptMetadata = {
@@ -336,11 +330,12 @@ result = task(..., run_in_background=false)  // Never wait synchronously for exp
 \`\`\`

 ### Background Result Collection:
-1. Launch parallel agents \u2192 receive task_ids
-2. Continue immediate work (explore, librarian results)
+1. Launch parallel agents → receive task_ids
+2. Continue immediate work
 3. When results needed: \`background_output(task_id="...")\`
-4. **If Oracle is running**: STOP all other output. Follow Oracle Completion Protocol in <Oracle_Usage>.
-5. Cleanup: Cancel disposable tasks (explore, librarian) individually via \`background_cancel(taskId="...")\`. Never use \`background_cancel(all=true)\`.
+4. Before final answer, cancel DISPOSABLE tasks (explore, librarian) individually: \`background_cancel(taskId="bg_explore_xxx")\`, \`background_cancel(taskId="bg_librarian_xxx")\`
+5. **NEVER cancel Oracle.** ALWAYS collect Oracle result via \`background_output(task_id="bg_oracle_xxx")\` before answering — even if you already have enough context.
+6. **NEVER use \`background_cancel(all=true)\`** — it kills Oracle. Cancel each disposable task by its specific taskId.

 ### Search Stop Conditions

@@ -477,9 +472,9 @@ If verification fails:
 3. Report: "Done. Note: found N pre-existing lint errors unrelated to my changes."

 ### Before Delivering Final Answer:
- **If Oracle is running**: STOP. Follow Oracle Completion Protocol in <Oracle_Usage>. Do NOT deliver any answer.
- Cancel disposable background tasks (explore, librarian) individually via \`background_cancel(taskId="...")\`.
- **Never use \`background_cancel(all=true)\`.**
+- Cancel DISPOSABLE background tasks (explore, librarian) individually via \`background_cancel(taskId="...")\`
+- **NEVER use \`background_cancel(all=true)\`.** Always cancel individually by taskId.
+- **Always wait for Oracle**: When Oracle is running and you have gathered enough context from your own exploration, your next action is \`background_output\` on Oracle — NOT delivering a final answer. Oracle's value is highest when you think you don't need it.
 </Behavior_Instructions>

 ${oracleSection}
@@ -553,7 +548,7 @@ export function createSisyphusAgent(
  const tools = availableToolNames ? categorizeTools(availableToolNames) : [];
  const skills = availableSkills ?? [];
  const categories = availableCategories ?? [];
-  let prompt = availableAgents
+  const prompt = availableAgents
    ? buildDynamicSisyphusPrompt(
        model,
        availableAgents,
@@ -564,15 +559,6 @@ export function createSisyphusAgent(
      )
    : buildDynamicSisyphusPrompt(model, [], tools, skills, categories, useTaskSystem);

-  if (isGeminiModel(model)) {
-    prompt = prompt.replace(
-      "</intent_verbalization>",
-      `</intent_verbalization>\n\n${buildGeminiIntentGateEnforcement()}\n\n${buildGeminiToolMandate()}`
-    );
-    prompt += "\n" + buildGeminiDelegationOverride();
-    prompt += "\n" + buildGeminiVerificationOverride();
-  }
-
  const permission = {
    question: "allow",
    call_omo_agent: "deny",
--- a/src/agents/tool-restrictions.test.ts
+++ b/src/agents/tool-restrictions.test.ts
@@ -4,7 +4,6 @@ import { createLibrarianAgent } from "./librarian"
 import { createExploreAgent } from "./explore"
 import { createMomusAgent } from "./momus"
 import { createMetisAgent } from "./metis"
-import { createAtlasAgent } from "./atlas"

 const TEST_MODEL = "anthropic/claude-sonnet-4-5"

@@ -97,18 +96,4 @@ describe("read-only agent tool restrictions", () => {
      }
    })
  })
-
-  describe("Atlas", () => {
-    test("allows delegation tools for orchestration", () => {
-      // given
-      const agent = createAtlasAgent({ model: TEST_MODEL })
-
-      // when
-      const permission = (agent.permission ?? {}) as Record<string, string>
-
-      // then
-      expect(permission["task"]).toBeUndefined()
-      expect(permission["call_omo_agent"]).toBeUndefined()
-    })
-  })
 })
--- a/src/agents/types.test.ts
+++ b/src/agents/types.test.ts
@@ -1,18 +1,12 @@
 import { describe, test, expect } from "bun:test";
-import { isGptModel, isGeminiModel } from "./types";
+import { isGptModel } from "./types";

 describe("isGptModel", () => {
-  test("standard openai provider gpt models", () => {
+  test("standard openai provider models", () => {
    expect(isGptModel("openai/gpt-5.2")).toBe(true);
    expect(isGptModel("openai/gpt-4o")).toBe(true);
-  });
-
-  test("o-series models are not gpt by name", () => {
-    expect(isGptModel("openai/o1")).toBe(false);
-    expect(isGptModel("openai/o3-mini")).toBe(false);
-    expect(isGptModel("litellm/o1")).toBe(false);
-    expect(isGptModel("litellm/o3-mini")).toBe(false);
-    expect(isGptModel("litellm/o4-mini")).toBe(false);
+    expect(isGptModel("openai/o1")).toBe(true);
+    expect(isGptModel("openai/o3-mini")).toBe(true);
  });

  test("github copilot gpt models", () => {
@@ -23,6 +17,9 @@ describe("isGptModel", () => {
  test("litellm proxied gpt models", () => {
    expect(isGptModel("litellm/gpt-5.2")).toBe(true);
    expect(isGptModel("litellm/gpt-4o")).toBe(true);
+    expect(isGptModel("litellm/o1")).toBe(true);
+    expect(isGptModel("litellm/o3-mini")).toBe(true);
+    expect(isGptModel("litellm/o4-mini")).toBe(true);
  });

  test("other proxied gpt models", () => {
@@ -30,11 +27,6 @@ describe("isGptModel", () => {
    expect(isGptModel("custom-provider/gpt-5.2")).toBe(true);
  });

-  test("venice provider gpt models", () => {
-    expect(isGptModel("venice/gpt-5.2")).toBe(true);
-    expect(isGptModel("venice/gpt-4o")).toBe(true);
-  });
-
  test("gpt4 prefix without hyphen (legacy naming)", () => {
    expect(isGptModel("litellm/gpt4o")).toBe(true);
    expect(isGptModel("ollama/gpt4")).toBe(true);
@@ -55,47 +47,3 @@ describe("isGptModel", () => {
    expect(isGptModel("opencode/claude-opus-4-6")).toBe(false);
  });
 });
-
-describe("isGeminiModel", () => {
-  test("#given google provider models #then returns true", () => {
-    expect(isGeminiModel("google/gemini-3-pro")).toBe(true);
-    expect(isGeminiModel("google/gemini-3-flash")).toBe(true);
-    expect(isGeminiModel("google/gemini-2.5-pro")).toBe(true);
-  });
-
-  test("#given google-vertex provider models #then returns true", () => {
-    expect(isGeminiModel("google-vertex/gemini-3-pro")).toBe(true);
-    expect(isGeminiModel("google-vertex/gemini-3-flash")).toBe(true);
-  });
-
-  test("#given github copilot gemini models #then returns true", () => {
-    expect(isGeminiModel("github-copilot/gemini-3-pro")).toBe(true);
-    expect(isGeminiModel("github-copilot/gemini-3-flash")).toBe(true);
-  });
-
-  test("#given litellm proxied gemini models #then returns true", () => {
-    expect(isGeminiModel("litellm/gemini-3-pro")).toBe(true);
-    expect(isGeminiModel("litellm/gemini-3-flash")).toBe(true);
-    expect(isGeminiModel("litellm/gemini-2.5-pro")).toBe(true);
-  });
-
-  test("#given other proxied gemini models #then returns true", () => {
-    expect(isGeminiModel("custom-provider/gemini-3-pro")).toBe(true);
-    expect(isGeminiModel("ollama/gemini-3-flash")).toBe(true);
-  });
-
-  test("#given gpt models #then returns false", () => {
-    expect(isGeminiModel("openai/gpt-5.2")).toBe(false);
-    expect(isGeminiModel("openai/o3-mini")).toBe(false);
-    expect(isGeminiModel("litellm/gpt-4o")).toBe(false);
-  });
-
-  test("#given claude models #then returns false", () => {
-    expect(isGeminiModel("anthropic/claude-opus-4-6")).toBe(false);
-    expect(isGeminiModel("anthropic/claude-sonnet-4-6")).toBe(false);
-  });
-
-  test("#given opencode provider #then returns false", () => {
-    expect(isGeminiModel("opencode/claude-opus-4-6")).toBe(false);
-  });
-});
--- a/src/agents/types.ts
+++ b/src/agents/types.ts
@@ -70,22 +70,14 @@ function extractModelName(model: string): string {
  return model.includes("/") ? model.split("/").pop() ?? model : model
 }

+const GPT_MODEL_PREFIXES = ["gpt-", "gpt4", "o1", "o3", "o4"]
+
 export function isGptModel(model: string): boolean {
-  const modelName = extractModelName(model).toLowerCase()
-  return modelName.includes("gpt")
-}
-
-const GEMINI_PROVIDERS = ["google/", "google-vertex/"]
-
-export function isGeminiModel(model: string): boolean {
-  if (GEMINI_PROVIDERS.some((prefix) => model.startsWith(prefix)))
-    return true
-
-  if (model.startsWith("github-copilot/") && extractModelName(model).toLowerCase().startsWith("gemini"))
+  if (model.startsWith("openai/") || model.startsWith("github-copilot/gpt-"))
    return true

  const modelName = extractModelName(model).toLowerCase()
-  return modelName.startsWith("gemini-")
+  return GPT_MODEL_PREFIXES.some((prefix) => modelName.startsWith(prefix))
 }

 export type BuiltinAgentName =
--- a/src/agents/utils.test.ts
+++ b/src/agents/utils.test.ts
@@ -589,22 +589,20 @@ describe("createBuiltinAgents with requiresProvider gating (hephaestus)", () =>
    }
  })

-  test("hephaestus IS created when github-copilot is connected with a GPT model", async () => {
-    // #given - github-copilot provider has gpt-5.3-codex available
+  test("hephaestus is created when github-copilot provider is connected", async () => {
+    // #given - github-copilot provider has models available
    const fetchSpy = spyOn(shared, "fetchAvailableModels").mockResolvedValue(
      new Set(["github-copilot/gpt-5.3-codex"])
    )
-    const cacheSpy = spyOn(connectedProvidersCache, "readConnectedProvidersCache").mockReturnValue(null)

    try {
      // #when
      const agents = await createBuiltinAgents([], {}, undefined, TEST_DEFAULT_MODEL, undefined, undefined, [], {})

-      // #then - github-copilot is now a valid provider for hephaestus
+      // #then
      expect(agents.hephaestus).toBeDefined()
    } finally {
      fetchSpy.mockRestore()
-      cacheSpy.mockRestore()
    }
  })

--- a/src/cli/AGENTS.md
+++ b/src/cli/AGENTS.md
@@ -1,6 +1,6 @@
 # src/cli/ — CLI: install, run, doctor, mcp-oauth

-**Generated:** 2026-02-24
+**Generated:** 2026-02-21

 ## OVERVIEW

--- a/src/cli/snapshots/model-fallback.test.ts.snap
+++ b/src/cli/snapshots/model-fallback.test.ts.snap
@@ -750,6 +750,10 @@ exports[`generateModelConfig fallback providers uses GitHub Copilot models when
    "explore": {
      "model": "github-copilot/gpt-5-mini",
    },
+    "hephaestus": {
+      "model": "github-copilot/gpt-5.3-codex",
+      "variant": "medium",
+    },
    "librarian": {
      "model": "github-copilot/claude-sonnet-4.5",
    },
@@ -782,12 +786,16 @@ exports[`generateModelConfig fallback providers uses GitHub Copilot models when
      "model": "github-copilot/gemini-3-pro-preview",
      "variant": "high",
    },
+    "deep": {
+      "model": "github-copilot/gpt-5.3-codex",
+      "variant": "medium",
+    },
    "quick": {
      "model": "github-copilot/claude-haiku-4.5",
    },
    "ultrabrain": {
-      "model": "github-copilot/gemini-3-pro-preview",
-      "variant": "high",
+      "model": "github-copilot/gpt-5.3-codex",
+      "variant": "xhigh",
    },
    "unspecified-high": {
      "model": "github-copilot/claude-sonnet-4.5",
@@ -816,6 +824,10 @@ exports[`generateModelConfig fallback providers uses GitHub Copilot models with
    "explore": {
      "model": "github-copilot/gpt-5-mini",
    },
+    "hephaestus": {
+      "model": "github-copilot/gpt-5.3-codex",
+      "variant": "medium",
+    },
    "librarian": {
      "model": "github-copilot/claude-sonnet-4.5",
    },
@@ -848,12 +860,16 @@ exports[`generateModelConfig fallback providers uses GitHub Copilot models with
      "model": "github-copilot/gemini-3-pro-preview",
      "variant": "high",
    },
+    "deep": {
+      "model": "github-copilot/gpt-5.3-codex",
+      "variant": "medium",
+    },
    "quick": {
      "model": "github-copilot/claude-haiku-4.5",
    },
    "ultrabrain": {
-      "model": "github-copilot/gemini-3-pro-preview",
-      "variant": "high",
+      "model": "github-copilot/gpt-5.3-codex",
+      "variant": "xhigh",
    },
    "unspecified-high": {
      "model": "github-copilot/claude-opus-4.6",
@@ -1269,7 +1285,7 @@ exports[`generateModelConfig mixed provider scenarios uses all fallback provider
      "model": "opencode/claude-haiku-4-5",
    },
    "hephaestus": {
-      "model": "opencode/gpt-5.3-codex",
+      "model": "github-copilot/gpt-5.3-codex",
      "variant": "medium",
    },
    "librarian": {
@@ -1305,14 +1321,14 @@ exports[`generateModelConfig mixed provider scenarios uses all fallback provider
      "variant": "high",
    },
    "deep": {
-      "model": "opencode/gpt-5.3-codex",
+      "model": "github-copilot/gpt-5.3-codex",
      "variant": "medium",
    },
    "quick": {
      "model": "github-copilot/claude-haiku-4.5",
    },
    "ultrabrain": {
-      "model": "opencode/gpt-5.3-codex",
+      "model": "github-copilot/gpt-5.3-codex",
      "variant": "xhigh",
    },
    "unspecified-high": {
--- a/src/cli/config-manager/AGENTS.md
+++ b/src/cli/config-manager/AGENTS.md
@@ -1,6 +1,6 @@
 # src/cli/config-manager/ — CLI Installation Utilities

-**Generated:** 2026-02-24
+**Generated:** 2026-02-21

 ## OVERVIEW

--- a/src/cli/config-manager/write-omo-config.test.ts
+++ b/src/cli/config-manager/write-omo-config.test.ts
@@ -1,80 +0,0 @@
-import { afterEach, beforeEach, describe, expect, it } from "bun:test"
-import { mkdirSync, readFileSync, rmSync, writeFileSync } from "node:fs"
-import { tmpdir } from "node:os"
-import { join } from "node:path"
-
-import { parseJsonc } from "../../shared/jsonc-parser"
-import type { InstallConfig } from "../types"
-import { resetConfigContext } from "./config-context"
-import { generateOmoConfig } from "./generate-omo-config"
-import { writeOmoConfig } from "./write-omo-config"
-
-const installConfig: InstallConfig = {
-  hasClaude: true,
-  isMax20: true,
-  hasOpenAI: true,
-  hasGemini: true,
-  hasCopilot: false,
-  hasOpencodeZen: false,
-  hasZaiCodingPlan: false,
-  hasKimiForCoding: false,
-}
-
-function getRecord(value: unknown): Record<string, unknown> {
-  if (value && typeof value === "object" && !Array.isArray(value)) {
-    return value as Record<string, unknown>
-  }
-
-  return {}
-}
-
-describe("writeOmoConfig", () => {
-  let testConfigDir = ""
-  let testConfigPath = ""
-
-  beforeEach(() => {
-    testConfigDir = join(tmpdir(), `omo-write-config-${Date.now()}-${Math.random().toString(36).slice(2)}`)
-    testConfigPath = join(testConfigDir, "oh-my-opencode.json")
-
-    mkdirSync(testConfigDir, { recursive: true })
-    process.env.OPENCODE_CONFIG_DIR = testConfigDir
-    resetConfigContext()
-  })
-
-  afterEach(() => {
-    rmSync(testConfigDir, { recursive: true, force: true })
-    resetConfigContext()
-    delete process.env.OPENCODE_CONFIG_DIR
-  })
-
-  it("preserves existing user values while adding new defaults", () => {
-    // given
-    const existingConfig = {
-      agents: {
-        sisyphus: {
-          model: "custom/provider-model",
-        },
-      },
-      disabled_hooks: ["comment-checker"],
-    }
-    writeFileSync(testConfigPath, JSON.stringify(existingConfig, null, 2) + "\n", "utf-8")
-
-    const generatedDefaults = generateOmoConfig(installConfig)
-
-    // when
-    const result = writeOmoConfig(installConfig)
-
-    // then
-    expect(result.success).toBe(true)
-
-    const savedConfig = parseJsonc<Record<string, unknown>>(readFileSync(testConfigPath, "utf-8"))
-    const savedAgents = getRecord(savedConfig.agents)
-    const savedSisyphus = getRecord(savedAgents.sisyphus)
-    expect(savedSisyphus.model).toBe("custom/provider-model")
-    expect(savedConfig.disabled_hooks).toEqual(["comment-checker"])
-
-    for (const defaultKey of Object.keys(generatedDefaults)) {
-      expect(savedConfig).toHaveProperty(defaultKey)
-    }
-  })
-})
--- a/src/cli/config-manager/write-omo-config.ts
+++ b/src/cli/config-manager/write-omo-config.ts
@@ -43,7 +43,7 @@ export function writeOmoConfig(installConfig: InstallConfig): ConfigMergeResult
          return { success: true, configPath: omoConfigPath }
        }

-        const merged = deepMergeRecord(newConfig, existing)
+        const merged = deepMergeRecord(existing, newConfig)
        writeFileSync(omoConfigPath, JSON.stringify(merged, null, 2) + "\n")
      } catch (parseErr) {
        if (parseErr instanceof SyntaxError) {
--- a/src/cli/doctor/formatter.test.ts
+++ b/src/cli/doctor/formatter.test.ts
@@ -1,5 +1,4 @@
-import { describe, expect, it } from "bun:test"
-import { stripAnsi } from "./format-shared"
+import { afterEach, describe, expect, it, mock } from "bun:test"
 import type { DoctorResult } from "./types"

 function createDoctorResult(): DoctorResult {
@@ -40,122 +39,78 @@ function createDoctorResult(): DoctorResult {
  }
 }

-function createDoctorResultWithIssues(): DoctorResult {
-  const base = createDoctorResult()
-  base.results[1].issues = [
-    { title: "Config issue", description: "Bad config", severity: "error" as const, fix: "Fix it" },
-    { title: "Tool warning", description: "Missing tool", severity: "warning" as const },
-  ]
-  base.summary.failed = 1
-  base.summary.warnings = 1
-  return base
-}
-
-describe("formatDoctorOutput", () => {
-  describe("#given default mode", () => {
-    it("shows System OK when no issues", async () => {
-      //#given
-      const result = createDoctorResult()
-      const { formatDoctorOutput } = await import(`./formatter?default-ok-${Date.now()}`)
-
-      //#when
-      const output = stripAnsi(formatDoctorOutput(result, "default"))
-
-      //#then
-      expect(output).toContain("System OK (opencode 1.0.200 · oh-my-opencode 3.4.0)")
-    })
-
-    it("shows issue count and details when issues exist", async () => {
-      //#given
-      const result = createDoctorResultWithIssues()
-      const { formatDoctorOutput } = await import(`./formatter?default-issues-${Date.now()}`)
-
-      //#when
-      const output = stripAnsi(formatDoctorOutput(result, "default"))
-
-      //#then
-      expect(output).toContain("issues found:")
-      expect(output).toContain("1. Config issue")
-      expect(output).toContain("2. Tool warning")
-    })
+describe("formatter", () => {
+  afterEach(() => {
+    mock.restore()
  })

-  describe("#given status mode", () => {
-    it("renders system version line", async () => {
+  describe("formatDoctorOutput", () => {
+    it("dispatches to default formatter for default mode", async () => {
      //#given
-      const result = createDoctorResult()
-      const { formatDoctorOutput } = await import(`./formatter?status-ver-${Date.now()}`)
+      const formatDefaultMock = mock(() => "default-output")
+      const formatStatusMock = mock(() => "status-output")
+      const formatVerboseMock = mock(() => "verbose-output")
+      mock.module("./format-default", () => ({ formatDefault: formatDefaultMock }))
+      mock.module("./format-status", () => ({ formatStatus: formatStatusMock }))
+      mock.module("./format-verbose", () => ({ formatVerbose: formatVerboseMock }))
+      const { formatDoctorOutput } = await import(`./formatter?default=${Date.now()}`)

      //#when
-      const output = stripAnsi(formatDoctorOutput(result, "status"))
+      const output = formatDoctorOutput(createDoctorResult(), "default")

      //#then
-      expect(output).toContain("1.0.200 · 3.4.0 · Bun 1.2.0")
+      expect(output).toBe("default-output")
+      expect(formatDefaultMock).toHaveBeenCalledTimes(1)
+      expect(formatStatusMock).toHaveBeenCalledTimes(0)
+      expect(formatVerboseMock).toHaveBeenCalledTimes(0)
    })

-    it("renders tool and MCP info", async () => {
+    it("dispatches to status formatter for status mode", async () => {
      //#given
-      const result = createDoctorResult()
-      const { formatDoctorOutput } = await import(`./formatter?status-tools-${Date.now()}`)
+      const formatDefaultMock = mock(() => "default-output")
+      const formatStatusMock = mock(() => "status-output")
+      const formatVerboseMock = mock(() => "verbose-output")
+      mock.module("./format-default", () => ({ formatDefault: formatDefaultMock }))
+      mock.module("./format-status", () => ({ formatStatus: formatStatusMock }))
+      mock.module("./format-verbose", () => ({ formatVerbose: formatVerboseMock }))
+      const { formatDoctorOutput } = await import(`./formatter?status=${Date.now()}`)

      //#when
-      const output = stripAnsi(formatDoctorOutput(result, "status"))
+      const output = formatDoctorOutput(createDoctorResult(), "status")

      //#then
-      expect(output).toContain("LSP 2/4")
-      expect(output).toContain("context7")
-    })
-  })
-
-  describe("#given verbose mode", () => {
-    it("includes all section headers", async () => {
-      //#given
-      const result = createDoctorResult()
-      const { formatDoctorOutput } = await import(`./formatter?verbose-headers-${Date.now()}`)
-
-      //#when
-      const output = stripAnsi(formatDoctorOutput(result, "verbose"))
-
-      //#then
-      expect(output).toContain("System Information")
-      expect(output).toContain("Configuration")
-      expect(output).toContain("Tools")
-      expect(output).toContain("MCPs")
-      expect(output).toContain("Summary")
+      expect(output).toBe("status-output")
+      expect(formatDefaultMock).toHaveBeenCalledTimes(0)
+      expect(formatStatusMock).toHaveBeenCalledTimes(1)
+      expect(formatVerboseMock).toHaveBeenCalledTimes(0)
    })

-    it("shows check summary counts", async () => {
+    it("dispatches to verbose formatter for verbose mode", async () => {
      //#given
-      const result = createDoctorResult()
-      const { formatDoctorOutput } = await import(`./formatter?verbose-summary-${Date.now()}`)
+      const formatDefaultMock = mock(() => "default-output")
+      const formatStatusMock = mock(() => "status-output")
+      const formatVerboseMock = mock(() => "verbose-output")
+      mock.module("./format-default", () => ({ formatDefault: formatDefaultMock }))
+      mock.module("./format-status", () => ({ formatStatus: formatStatusMock }))
+      mock.module("./format-verbose", () => ({ formatVerbose: formatVerboseMock }))
+      const { formatDoctorOutput } = await import(`./formatter?verbose=${Date.now()}`)

      //#when
-      const output = stripAnsi(formatDoctorOutput(result, "verbose"))
+      const output = formatDoctorOutput(createDoctorResult(), "verbose")

      //#then
-      expect(output).toContain("1 passed")
-      expect(output).toContain("0 failed")
-      expect(output).toContain("1 warnings")
+      expect(output).toBe("verbose-output")
+      expect(formatDefaultMock).toHaveBeenCalledTimes(0)
+      expect(formatStatusMock).toHaveBeenCalledTimes(0)
+      expect(formatVerboseMock).toHaveBeenCalledTimes(1)
    })
  })

  describe("formatJsonOutput", () => {
-    it("returns valid JSON", async () => {
+    it("returns valid JSON payload", async () => {
      //#given
+      const { formatJsonOutput } = await import(`./formatter?json=${Date.now()}`)
      const result = createDoctorResult()
-      const { formatJsonOutput } = await import(`./formatter?json-valid-${Date.now()}`)
-
-      //#when
-      const output = formatJsonOutput(result)
-
-      //#then
-      expect(() => JSON.parse(output)).not.toThrow()
-    })
-
-    it("preserves all result fields", async () => {
-      //#given
-      const result = createDoctorResult()
-      const { formatJsonOutput } = await import(`./formatter?json-fields-${Date.now()}`)

      //#when
      const output = formatJsonOutput(result)
@@ -164,6 +119,7 @@ describe("formatDoctorOutput", () => {
      //#then
      expect(parsed.summary.total).toBe(2)
      expect(parsed.systemInfo.pluginVersion).toBe("3.4.0")
+      expect(parsed.tools.ghCli.username).toBe("yeongyu")
      expect(parsed.exitCode).toBe(0)
    })
  })
--- a/src/cli/model-fallback-requirements.ts
+++ b/src/cli/model-fallback-requirements.ts
@@ -17,9 +17,9 @@ export const CLI_AGENT_MODEL_REQUIREMENTS: Record<string, ModelRequirement> = {
  },
  hephaestus: {
    fallbackChain: [
-      { providers: ["openai", "opencode"], model: "gpt-5.3-codex", variant: "medium" },
+      { providers: ["openai", "github-copilot", "opencode"], model: "gpt-5.3-codex", variant: "medium" },
    ],
-    requiresProvider: ["openai", "opencode"],
+    requiresProvider: ["openai", "github-copilot", "opencode"],
  },
  oracle: {
    fallbackChain: [
@@ -100,14 +100,14 @@ export const CLI_CATEGORY_MODEL_REQUIREMENTS: Record<string, ModelRequirement> =
  },
  ultrabrain: {
    fallbackChain: [
-      { providers: ["openai", "opencode"], model: "gpt-5.3-codex", variant: "xhigh" },
+      { providers: ["openai", "github-copilot", "opencode"], model: "gpt-5.3-codex", variant: "xhigh" },
      { providers: ["google", "github-copilot", "opencode"], model: "gemini-3-pro", variant: "high" },
      { providers: ["anthropic", "github-copilot", "opencode"], model: "claude-opus-4-6", variant: "max" },
    ],
  },
  deep: {
    fallbackChain: [
-      { providers: ["openai", "opencode"], model: "gpt-5.3-codex", variant: "medium" },
+      { providers: ["openai", "github-copilot", "opencode"], model: "gpt-5.3-codex", variant: "medium" },
      { providers: ["anthropic", "github-copilot", "opencode"], model: "claude-opus-4-6", variant: "max" },
      { providers: ["google", "github-copilot", "opencode"], model: "gemini-3-pro", variant: "high" },
    ],
@@ -131,7 +131,7 @@ export const CLI_CATEGORY_MODEL_REQUIREMENTS: Record<string, ModelRequirement> =
  "unspecified-low": {
    fallbackChain: [
      { providers: ["anthropic", "github-copilot", "opencode"], model: "claude-sonnet-4-5" },
-      { providers: ["openai", "opencode"], model: "gpt-5.3-codex", variant: "medium" },
+      { providers: ["openai", "github-copilot", "opencode"], model: "gpt-5.3-codex", variant: "medium" },
      { providers: ["google", "github-copilot", "opencode"], model: "gemini-3-flash" },
    ],
  },
--- a/src/cli/model-fallback.test.ts
+++ b/src/cli/model-fallback.test.ts
@@ -421,15 +421,16 @@ describe("generateModelConfig", () => {
      expect(result.agents?.hephaestus?.variant).toBe("medium")
    })

-    test("Hephaestus is NOT created when only Copilot is available (gpt-5.3-codex unavailable on github-copilot)", () => {
+    test("Hephaestus is created when Copilot is available (github-copilot provider connected)", () => {
      // #given
      const config = createConfig({ hasCopilot: true })

      // #when
      const result = generateModelConfig(config)

-      // #then - hephaestus is omitted because gpt-5.3-codex is not available on github-copilot
-      expect(result.agents?.hephaestus).toBeUndefined()
+      // #then
+      expect(result.agents?.hephaestus?.model).toBe("github-copilot/gpt-5.3-codex")
+      expect(result.agents?.hephaestus?.variant).toBe("medium")
    })

    test("Hephaestus is created when OpenCode Zen is available (opencode provider connected)", () => {
--- a/src/cli/run/AGENTS.md
+++ b/src/cli/run/AGENTS.md
@@ -1,6 +1,6 @@
 # src/cli/run/ — Non-Interactive Session Launcher

-**Generated:** 2026-02-24
+**Generated:** 2026-02-21

 ## OVERVIEW

--- a/src/cli/run/event-handlers.test.ts
+++ b/src/cli/run/event-handlers.test.ts
@@ -1,4 +1,4 @@
-const { describe, it, expect, spyOn } = require("bun:test")
+import { describe, it, expect, spyOn } from "bun:test"
 import type { RunContext } from "./types"
 import { createEventState } from "./events"
 import { handleSessionStatus, handleMessagePartUpdated, handleMessageUpdated, handleTuiToast } from "./event-handlers"
@@ -235,7 +235,9 @@ describe("handleMessagePartUpdated", () => {

  it("prints completion metadata once when assistant text part is completed", () => {
    // given
-    const nowSpy = spyOn(Date, "now").mockReturnValue(3400)
+    const nowSpy = spyOn(Date, "now")
+    nowSpy.mockReturnValueOnce(1000)
+    nowSpy.mockReturnValueOnce(3400)

    const ctx = createMockContext("ses_main")
    const state = createEventState()
@@ -257,7 +259,6 @@ describe("handleMessagePartUpdated", () => {
      } as any,
      state,
    )
-    state.messageStartedAtById["msg_1"] = 1000

    // when
    handleMessagePartUpdated(
--- a/src/cli/run/event-state.ts
+++ b/src/cli/run/event-state.ts
@@ -7,8 +7,6 @@ export interface EventState {
  currentTool: string | null
  /** Set to true when the main session has produced meaningful work (text, tool call, or tool result) */
  hasReceivedMeaningfulWork: boolean
-  /** Timestamp of the last received event (for watchdog detection) */
-  lastEventTimestamp: number
  /** Count of assistant messages for the main session */
  messageCount: number
  /** Current agent name from the latest assistant message */
@@ -56,7 +54,6 @@ export function createEventState(): EventState {
    lastPartText: "",
    currentTool: null,
    hasReceivedMeaningfulWork: false,
-    lastEventTimestamp: Date.now(),
    messageCount: 0,
    currentAgent: null,
    currentModel: null,
--- a/src/cli/run/event-stream-processor.ts
+++ b/src/cli/run/event-stream-processor.ts
@@ -35,9 +35,6 @@ export async function processEvents(
        logEventVerbose(ctx, payload)
      }

-      // Update last event timestamp for watchdog detection
-      state.lastEventTimestamp = Date.now()
-
      handleSessionError(ctx, payload, state)
      handleSessionIdle(ctx, payload, state)
      handleSessionStatus(ctx, payload, state)
--- a/src/cli/run/poll-for-completion.ts
+++ b/src/cli/run/poll-for-completion.ts
@@ -8,15 +8,11 @@ const DEFAULT_POLL_INTERVAL_MS = 500
 const DEFAULT_REQUIRED_CONSECUTIVE = 1
 const ERROR_GRACE_CYCLES = 3
 const MIN_STABILIZATION_MS = 1_000
-const DEFAULT_EVENT_WATCHDOG_MS = 30_000 // 30 seconds
-const DEFAULT_SECONDARY_MEANINGFUL_WORK_TIMEOUT_MS = 60_000 // 60 seconds

 export interface PollOptions {
  pollIntervalMs?: number
  requiredConsecutive?: number
  minStabilizationMs?: number
-  eventWatchdogMs?: number
-  secondaryMeaningfulWorkTimeoutMs?: number
 }

 export async function pollForCompletion(
@@ -32,15 +28,9 @@ export async function pollForCompletion(
    options.minStabilizationMs ?? MIN_STABILIZATION_MS
  const minStabilizationMs =
    rawMinStabilizationMs > 0 ? rawMinStabilizationMs : MIN_STABILIZATION_MS
-  const eventWatchdogMs =
-    options.eventWatchdogMs ?? DEFAULT_EVENT_WATCHDOG_MS
-  const secondaryMeaningfulWorkTimeoutMs =
-    options.secondaryMeaningfulWorkTimeoutMs ??
-    DEFAULT_SECONDARY_MEANINGFUL_WORK_TIMEOUT_MS
  let consecutiveCompleteChecks = 0
  let errorCycleCount = 0
  let firstWorkTimestamp: number | null = null
-  let secondaryTimeoutChecked = false
  const pollStartTimestamp = Date.now()

  while (!abortController.signal.aborted) {
@@ -69,37 +59,7 @@ export async function pollForCompletion(
      errorCycleCount = 0
    }

-    // Watchdog: if no events received for N seconds, verify session status via API
-    let mainSessionStatus: "idle" | "busy" | "retry" | null = null
-    if (eventState.lastEventTimestamp !== null) {
-      const timeSinceLastEvent = Date.now() - eventState.lastEventTimestamp
-      if (timeSinceLastEvent > eventWatchdogMs) {
-        // Events stopped coming - verify actual session state
-        console.log(
-          pc.yellow(
-            `\n  No events for ${Math.round(
-              timeSinceLastEvent / 1000
-            )}s, verifying session status...`
-          )
-        )
-
-        // Force check session status directly
-        mainSessionStatus = await getMainSessionStatus(ctx)
-        if (mainSessionStatus === "idle") {
-          eventState.mainSessionIdle = true
-        } else if (mainSessionStatus === "busy" || mainSessionStatus === "retry") {
-          eventState.mainSessionIdle = false
-        }
-
-        // Reset timestamp to avoid repeated checks
-        eventState.lastEventTimestamp = Date.now()
-      }
-    }
-
-    // Only call getMainSessionStatus if watchdog didn't already check
-    if (mainSessionStatus === null) {
-      mainSessionStatus = await getMainSessionStatus(ctx)
-    }
+    const mainSessionStatus = await getMainSessionStatus(ctx)
    if (mainSessionStatus === "busy" || mainSessionStatus === "retry") {
      eventState.mainSessionIdle = false
    } else if (mainSessionStatus === "idle") {
@@ -121,50 +81,6 @@ export async function pollForCompletion(
        consecutiveCompleteChecks = 0
        continue
      }
-
-      // Secondary timeout: if we've been polling for reasonable time but haven't
-      // received meaningful work via events, check if there's active work via API
-      // Only check once to avoid unnecessary API calls every poll cycle
-      if (
-        Date.now() - pollStartTimestamp > secondaryMeaningfulWorkTimeoutMs &&
-        !secondaryTimeoutChecked
-      ) {
-        secondaryTimeoutChecked = true
-        // Check if session actually has pending work (children, todos, etc.)
-        const childrenRes = await ctx.client.session.children({
-          path: { id: ctx.sessionID },
-          query: { directory: ctx.directory },
-        })
-        const children = normalizeSDKResponse(childrenRes, [] as unknown[])
-        const todosRes = await ctx.client.session.todo({
-          path: { id: ctx.sessionID },
-          query: { directory: ctx.directory },
-        })
-        const todos = normalizeSDKResponse(todosRes, [] as unknown[])
-
-        const hasActiveChildren =
-          Array.isArray(children) && children.length > 0
-        const hasActiveTodos =
-          Array.isArray(todos) &&
-          todos.some(
-            (t: unknown) =>
-              (t as { status?: string })?.status !== "completed" &&
-              (t as { status?: string })?.status !== "cancelled"
-          )
-        const hasActiveWork = hasActiveChildren || hasActiveTodos
-
-        if (hasActiveWork) {
-          // Assume meaningful work is happening even without events
-          eventState.hasReceivedMeaningfulWork = true
-          console.log(
-            pc.yellow(
-              `\n  No meaningful work events for ${Math.round(
-                secondaryMeaningfulWorkTimeoutMs / 1000
-              )}s but session has active work - assuming in progress`
-            )
-          )
-        }
-      }
    } else {
      // Track when first meaningful work was received
      if (firstWorkTimestamp === null) {
--- a/src/cli/run/session-resolver.ts
+++ b/src/cli/run/session-resolver.ts
@@ -31,7 +31,7 @@ export async function resolveSession(options: {
        permission: [
          { permission: "question", action: "deny" as const, pattern: "*" },
        ],
-      } as Record<string, unknown>,
+      } as any,
      query: { directory },
    })

--- a/src/config/AGENTS.md
+++ b/src/config/AGENTS.md
@@ -1,6 +1,6 @@
 # src/config/ — Zod v4 Schema System

-**Generated:** 2026-02-24
+**Generated:** 2026-02-21

 ## OVERVIEW

--- a/src/config/index.ts
+++ b/src/config/index.ts
@@ -1,5 +1,18 @@
 export {
  OhMyOpenCodeConfigSchema,
+  AgentOverrideConfigSchema,
+  AgentOverridesSchema,
+  McpNameSchema,
+  AgentNameSchema,
+  HookNameSchema,
+  BuiltinCommandNameSchema,
+  SisyphusAgentConfigSchema,
+  ExperimentalConfigSchema,
+  RalphLoopConfigSchema,
+  TmuxConfigSchema,
+  TmuxLayoutSchema,
+  RuntimeFallbackConfigSchema,
+  FallbackModelsSchema,
 } from "./schema"

 export type {
--- a/src/config/schema/agent-overrides.ts
+++ b/src/config/schema/agent-overrides.ts
@@ -47,21 +47,13 @@ export const AgentOverrideConfigSchema = z.object({
      variant: z.string().optional(),
    })
    .optional(),
-  compaction: z
-    .object({
-      model: z.string().optional(),
-      variant: z.string().optional(),
-    })
-    .optional(),
 })

 export const AgentOverridesSchema = z.object({
  build: AgentOverrideConfigSchema.optional(),
  plan: AgentOverrideConfigSchema.optional(),
  sisyphus: AgentOverrideConfigSchema.optional(),
-  hephaestus: AgentOverrideConfigSchema.extend({
-    allow_non_gpt_model: z.boolean().optional(),
-  }).optional(),
+  hephaestus: AgentOverrideConfigSchema.optional(),
  "sisyphus-junior": AgentOverrideConfigSchema.optional(),
  "OpenCode-Builder": AgentOverrideConfigSchema.optional(),
  prometheus: AgentOverrideConfigSchema.optional(),
--- a/src/config/schema/categories.ts
+++ b/src/config/schema/categories.ts
@@ -20,7 +20,6 @@ export const CategoryConfigSchema = z.object({
  textVerbosity: z.enum(["low", "medium", "high"]).optional(),
  tools: z.record(z.string(), z.boolean()).optional(),
  prompt_append: z.string().optional(),
-  max_prompt_tokens: z.number().int().positive().optional(),
  /** Mark agent as unstable - forces background mode for monitoring. Auto-enabled for gemini/minimax models. */
  is_unstable_agent: z.boolean().optional(),
  /** Disable this category. Disabled categories are excluded from task delegation. */
--- a/src/config/schema/oh-my-opencode-config.ts
+++ b/src/config/schema/oh-my-opencode-config.ts
@@ -35,8 +35,6 @@ export const OhMyOpenCodeConfigSchema = z.object({
  disabled_tools: z.array(z.string()).optional(),
  /** Enable hashline_edit tool/hook integrations (default: true at call site) */
  hashline_edit: z.boolean().optional(),
-  /** Enable model fallback on API errors (default: false). Set to true to enable automatic model switching when model errors occur. */
-  model_fallback: z.boolean().optional(),
  agents: AgentOverridesSchema.optional(),
  categories: CategoriesConfigSchema.optional(),
  claude_code: ClaudeCodeConfigSchema.optional(),
--- a/src/features/AGENTS.md
+++ b/src/features/AGENTS.md
@@ -1,6 +1,6 @@
 # src/features/ — 19 Feature Modules

-**Generated:** 2026-02-24
+**Generated:** 2026-02-21

 ## OVERVIEW

--- a/src/features/background-agent/AGENTS.md
+++ b/src/features/background-agent/AGENTS.md
@@ -1,6 +1,6 @@
 # src/features/background-agent/ — Core Orchestration Engine

-**Generated:** 2026-02-24
+**Generated:** 2026-02-21

 ## OVERVIEW

--- a/src/features/background-agent/background-task-completer.ts
+++ b/src/features/background-agent/background-task-completer.ts
@@ -0,0 +1,40 @@
+import type { BackgroundTask } from "./types"
+import type { ResultHandlerContext } from "./result-handler-context"
+import { log } from "../../shared"
+import { notifyParentSession } from "./parent-session-notifier"
+
+export async function tryCompleteTask(
+  task: BackgroundTask,
+  source: string,
+  ctx: ResultHandlerContext
+): Promise<boolean> {
+  const { concurrencyManager, state } = ctx
+
+  if (task.status !== "running") {
+    log("[background-agent] Task already completed, skipping:", {
+      taskId: task.id,
+      status: task.status,
+      source,
+    })
+    return false
+  }
+
+  task.status = "completed"
+  task.completedAt = new Date()
+
+  if (task.concurrencyKey) {
+    concurrencyManager.release(task.concurrencyKey)
+    task.concurrencyKey = undefined
+  }
+
+  state.markForNotification(task)
+
+  try {
+    await notifyParentSession(task, ctx)
+    log(`[background-agent] Task completed via ${source}:`, task.id)
+  } catch (error) {
+    log("[background-agent] Error in notifyParentSession:", { taskId: task.id, error })
+  }
+
+  return true
+}
--- a/src/features/background-agent/format-duration.ts
+++ b/src/features/background-agent/format-duration.ts
@@ -0,0 +1,14 @@
+export function formatDuration(start: Date, end?: Date): string {
+  const duration = (end ?? new Date()).getTime() - start.getTime()
+  const seconds = Math.floor(duration / 1000)
+  const minutes = Math.floor(seconds / 60)
+  const hours = Math.floor(minutes / 60)
+
+  if (hours > 0) {
+    return `${hours}h ${minutes % 60}m ${seconds % 60}s`
+  }
+  if (minutes > 0) {
+    return `${minutes}m ${seconds % 60}s`
+  }
+  return `${seconds}s`
+}
--- a/src/features/background-agent/index.ts
+++ b/src/features/background-agent/index.ts
@@ -1,2 +1,5 @@
 export * from "./types"
 export { BackgroundManager, type SubagentSessionCreatedEvent, type OnSubagentSessionCreated } from "./manager"
+export { TaskHistory, type TaskHistoryEntry } from "./task-history"
+export { ConcurrencyManager } from "./concurrency"
+export { TaskStateManager } from "./state"
--- a/src/features/background-agent/manager.test.ts
+++ b/src/features/background-agent/manager.test.ts
@@ -191,10 +191,6 @@ function getPendingByParent(manager: BackgroundManager): Map<string, Set<string>
  return (manager as unknown as { pendingByParent: Map<string, Set<string>> }).pendingByParent
 }

-function getPendingNotifications(manager: BackgroundManager): Map<string, string[]> {
-  return (manager as unknown as { pendingNotifications: Map<string, string[]> }).pendingNotifications
-}
-
 function getCompletionTimers(manager: BackgroundManager): Map<string, ReturnType<typeof setTimeout>> {
  return (manager as unknown as { completionTimers: Map<string, ReturnType<typeof setTimeout>> }).completionTimers
 }
@@ -1061,49 +1057,6 @@ describe("BackgroundManager.notifyParentSession - aborted parent", () => {

    manager.shutdown()
  })
-
-  test("should queue notification when promptAsync aborts while parent is idle", async () => {
-    //#given
-    const promptMock = async () => {
-      const error = new Error("Request aborted while waiting for input")
-      error.name = "MessageAbortedError"
-      throw error
-    }
-    const client = {
-      session: {
-        prompt: promptMock,
-        promptAsync: promptMock,
-        abort: async () => ({}),
-        messages: async () => ({ data: [] }),
-      },
-    }
-    const manager = new BackgroundManager({ client, directory: tmpdir() } as unknown as PluginInput)
-    const task: BackgroundTask = {
-      id: "task-aborted-idle-queue",
-      sessionID: "session-child",
-      parentSessionID: "session-parent",
-      parentMessageID: "msg-parent",
-      description: "task idle queue",
-      prompt: "test",
-      agent: "explore",
-      status: "completed",
-      startedAt: new Date(),
-      completedAt: new Date(),
-    }
-    getPendingByParent(manager).set("session-parent", new Set([task.id]))
-
-    //#when
-    await (manager as unknown as { notifyParentSession: (task: BackgroundTask) => Promise<void> })
-      .notifyParentSession(task)
-
-    //#then
-    const queuedNotifications = getPendingNotifications(manager).get("session-parent") ?? []
-    expect(queuedNotifications).toHaveLength(1)
-    expect(queuedNotifications[0]).toContain("<system-reminder>")
-    expect(queuedNotifications[0]).toContain("[ALL BACKGROUND TASKS COMPLETE]")
-
-    manager.shutdown()
-  })
 })

 describe("BackgroundManager.notifyParentSession - notifications toggle", () => {
@@ -1152,29 +1105,6 @@ describe("BackgroundManager.notifyParentSession - notifications toggle", () => {
  })
 })

-describe("BackgroundManager.injectPendingNotificationsIntoChatMessage", () => {
-  test("should prepend queued notifications to first text part and clear queue", () => {
-    // given
-    const manager = createBackgroundManager()
-    manager.queuePendingNotification("session-parent", "<system-reminder>queued-one</system-reminder>")
-    manager.queuePendingNotification("session-parent", "<system-reminder>queued-two</system-reminder>")
-    const output = {
-      parts: [{ type: "text", text: "User prompt" }],
-    }
-
-    // when
-    manager.injectPendingNotificationsIntoChatMessage(output, "session-parent")
-
-    // then
-    expect(output.parts[0].text).toContain("<system-reminder>queued-one</system-reminder>")
-    expect(output.parts[0].text).toContain("<system-reminder>queued-two</system-reminder>")
-    expect(output.parts[0].text).toContain("User prompt")
-    expect(getPendingNotifications(manager).get("session-parent")).toBeUndefined()
-
-    manager.shutdown()
-  })
-})
-
 function buildNotificationPromptBody(
  task: BackgroundTask,
  currentMessage: CurrentMessage | null
--- a/src/features/background-agent/manager.ts
+++ b/src/features/background-agent/manager.ts
@@ -25,6 +25,7 @@ import {
  hasMoreFallbacks,
 } from "../../shared/model-error-classifier"
 import {
+  MIN_IDLE_TIME_MS,
  POLLING_INTERVAL_MS,
  TASK_CLEANUP_DELAY_MS,
 } from "./constants"
@@ -42,7 +43,6 @@ import {
 import { tryFallbackRetry } from "./fallback-retry-handler"
 import { registerManagerForCleanup, unregisterManagerForCleanup } from "./process-cleanup"
 import { isCompactionAgent, findNearestMessageExcludingCompaction } from "./compaction-aware-message-resolver"
-import { handleSessionIdleBackgroundEvent } from "./session-idle-event-handler"
 import { MESSAGE_STORAGE } from "../hook-message-injector"
 import { join } from "node:path"
 import { pruneStaleTasksAndNotifications } from "./task-poller"
@@ -93,7 +93,6 @@ export class BackgroundManager {

  private tasks: Map<string, BackgroundTask>
  private notifications: Map<string, BackgroundTask[]>
-  private pendingNotifications: Map<string, string[]>
  private pendingByParent: Map<string, Set<string>>  // Track pending tasks per parent for batching
  private client: OpencodeClient
  private directory: string
@@ -126,7 +125,6 @@ export class BackgroundManager {
  ) {
    this.tasks = new Map()
    this.notifications = new Map()
-    this.pendingNotifications = new Map()
    this.pendingByParent = new Map()
    this.client = ctx.client
    this.directory = ctx.directory
@@ -270,7 +268,7 @@ export class BackgroundManager {
      body: {
        parentID: input.parentSessionID,
        title: `${input.description} (@${input.agent} subagent)`,
-      } as Record<string, unknown>,
+      } as any,
      query: {
        directory: parentDirectory,
      },
@@ -742,15 +740,61 @@ export class BackgroundManager {
    }

    if (event.type === "session.idle") {
-      if (!props || typeof props !== "object") return
-      handleSessionIdleBackgroundEvent({
-        properties: props as Record<string, unknown>,
-        findBySession: (id) => this.findBySession(id),
-        idleDeferralTimers: this.idleDeferralTimers,
-        validateSessionHasOutput: (id) => this.validateSessionHasOutput(id),
-        checkSessionTodos: (id) => this.checkSessionTodos(id),
-        tryCompleteTask: (task, source) => this.tryCompleteTask(task, source),
-        emitIdleEvent: (sessionID) => this.handleEvent({ type: "session.idle", properties: { sessionID } }),
+      const sessionID = props?.sessionID as string | undefined
+      if (!sessionID) return
+
+      const task = this.findBySession(sessionID)
+      if (!task || task.status !== "running") return
+      
+      const startedAt = task.startedAt
+      if (!startedAt) return
+
+      // Edge guard: Require minimum elapsed time (5 seconds) before accepting idle
+      const elapsedMs = Date.now() - startedAt.getTime()
+      if (elapsedMs < MIN_IDLE_TIME_MS) {
+        const remainingMs = MIN_IDLE_TIME_MS - elapsedMs
+        if (!this.idleDeferralTimers.has(task.id)) {
+          log("[background-agent] Deferring early session.idle:", { elapsedMs, remainingMs, taskId: task.id })
+          const timer = setTimeout(() => {
+            this.idleDeferralTimers.delete(task.id)
+            this.handleEvent({ type: "session.idle", properties: { sessionID } })
+          }, remainingMs)
+          this.idleDeferralTimers.set(task.id, timer)
+        } else {
+          log("[background-agent] session.idle already deferred:", { elapsedMs, taskId: task.id })
+        }
+        return
+      }
+
+      // Edge guard: Verify session has actual assistant output before completing
+      this.validateSessionHasOutput(sessionID).then(async (hasValidOutput) => {
+        // Re-check status after async operation (could have been completed by polling)
+        if (task.status !== "running") {
+          log("[background-agent] Task status changed during validation, skipping:", { taskId: task.id, status: task.status })
+          return
+        }
+
+        if (!hasValidOutput) {
+          log("[background-agent] Session.idle but no valid output yet, waiting:", task.id)
+          return
+        }
+
+        const hasIncompleteTodos = await this.checkSessionTodos(sessionID)
+
+        // Re-check status after async operation again
+        if (task.status !== "running") {
+          log("[background-agent] Task status changed during todo check, skipping:", { taskId: task.id, status: task.status })
+          return
+        }
+
+        if (hasIncompleteTodos) {
+          log("[background-agent] Task has incomplete todos, waiting for todo-continuation:", task.id)
+          return
+        }
+
+        await this.tryCompleteTask(task, "session.idle event")
+      }).catch(err => {
+        log("[background-agent] Error in session.idle handler:", err)
      })
    }

@@ -919,32 +963,6 @@ export class BackgroundManager {
    this.notifications.delete(sessionID)
  }

-  queuePendingNotification(sessionID: string | undefined, notification: string): void {
-    if (!sessionID) return
-    const existingNotifications = this.pendingNotifications.get(sessionID) ?? []
-    existingNotifications.push(notification)
-    this.pendingNotifications.set(sessionID, existingNotifications)
-  }
-
-  injectPendingNotificationsIntoChatMessage(output: { parts: Array<{ type: string; text?: string; [key: string]: unknown }> }, sessionID: string): void {
-    const pendingNotifications = this.pendingNotifications.get(sessionID)
-    if (!pendingNotifications || pendingNotifications.length === 0) {
-      return
-    }
-
-    this.pendingNotifications.delete(sessionID)
-    const notificationContent = pendingNotifications.join("\n\n")
-    const firstTextPartIndex = output.parts.findIndex((part) => part.type === "text")
-
-    if (firstTextPartIndex === -1) {
-      output.parts.unshift(createInternalAgentTextPart(notificationContent))
-      return
-    }
-
-    const originalText = output.parts[firstTextPartIndex].text ?? ""
-    output.parts[firstTextPartIndex].text = `${notificationContent}\n\n---\n\n${originalText}`
-  }
-
  /**
   * Validates that a session has actual assistant/tool output before marking complete.
   * Prevents premature completion when session.idle fires before agent responds.
@@ -1368,7 +1386,6 @@ Use \`background_output(task_id="${task.id}")\` to retrieve this result when rea
              taskId: task.id,
              parentSessionID: task.parentSessionID,
            })
-            this.queuePendingNotification(task.parentSessionID, notification)
          } else {
            log("[background-agent] Failed to send notification:", error)
          }
@@ -1597,7 +1614,6 @@ Use \`background_output(task_id="${task.id}")\` to retrieve this result when rea
    this.concurrencyManager.clear()
    this.tasks.clear()
    this.notifications.clear()
-    this.pendingNotifications.clear()
    this.pendingByParent.clear()
    this.notificationQueueByParent.clear()
    this.queuesByKey.clear()
--- a/src/features/background-agent/message-dir.ts
+++ b/src/features/background-agent/message-dir.ts
@@ -0,0 +1 @@
+export { getMessageDir } from "../../shared"
--- a/src/features/background-agent/parent-session-context-resolver.ts
+++ b/src/features/background-agent/parent-session-context-resolver.ts
@@ -0,0 +1,81 @@
+import type { OpencodeClient } from "./constants"
+import type { BackgroundTask } from "./types"
+import { findNearestMessageWithFields } from "../hook-message-injector"
+import { getMessageDir } from "../../shared"
+import { normalizePromptTools, resolveInheritedPromptTools } from "../../shared"
+
+type AgentModel = { providerID: string; modelID: string }
+
+function isObject(value: unknown): value is Record<string, unknown> {
+  return typeof value === "object" && value !== null
+}
+
+function extractAgentAndModelFromMessage(message: unknown): {
+  agent?: string
+  model?: AgentModel
+  tools?: Record<string, boolean>
+} {
+  if (!isObject(message)) return {}
+  const info = message["info"]
+  if (!isObject(info)) return {}
+
+  const agent = typeof info["agent"] === "string" ? info["agent"] : undefined
+  const modelObj = info["model"]
+  const tools = normalizePromptTools(isObject(info["tools"]) ? info["tools"] as Record<string, unknown> as Record<string, boolean | "allow" | "deny" | "ask"> : undefined)
+  if (isObject(modelObj)) {
+    const providerID = modelObj["providerID"]
+    const modelID = modelObj["modelID"]
+    if (typeof providerID === "string" && typeof modelID === "string") {
+      return { agent, model: { providerID, modelID }, tools }
+    }
+  }
+
+  const providerID = info["providerID"]
+  const modelID = info["modelID"]
+  if (typeof providerID === "string" && typeof modelID === "string") {
+    return { agent, model: { providerID, modelID }, tools }
+  }
+
+  return { agent, tools }
+}
+
+export async function resolveParentSessionAgentAndModel(input: {
+  client: OpencodeClient
+  task: BackgroundTask
+}): Promise<{ agent?: string; model?: AgentModel; tools?: Record<string, boolean> }> {
+  const { client, task } = input
+
+  let agent: string | undefined = task.parentAgent
+  let model: AgentModel | undefined
+  let tools: Record<string, boolean> | undefined = task.parentTools
+
+  try {
+    const messagesResp = await client.session.messages({
+      path: { id: task.parentSessionID },
+    })
+
+    const messagesRaw = "data" in messagesResp ? messagesResp.data : []
+    const messages = Array.isArray(messagesRaw) ? messagesRaw : []
+
+    for (let i = messages.length - 1; i >= 0; i--) {
+      const extracted = extractAgentAndModelFromMessage(messages[i])
+      if (extracted.agent || extracted.model || extracted.tools) {
+        agent = extracted.agent ?? task.parentAgent
+        model = extracted.model
+        tools = extracted.tools ?? tools
+        break
+      }
+    }
+  } catch {
+    const messageDir = getMessageDir(task.parentSessionID)
+    const currentMessage = messageDir ? findNearestMessageWithFields(messageDir) : null
+    agent = currentMessage?.agent ?? task.parentAgent
+    model =
+      currentMessage?.model?.providerID && currentMessage?.model?.modelID
+        ? { providerID: currentMessage.model.providerID, modelID: currentMessage.model.modelID }
+        : undefined
+    tools = normalizePromptTools(currentMessage?.tools) ?? tools
+  }
+
+  return { agent, model, tools: resolveInheritedPromptTools(task.parentSessionID, tools) }
+}
--- a/src/features/background-agent/parent-session-notifier.test.ts
+++ b/src/features/background-agent/parent-session-notifier.test.ts
@@ -0,0 +1,39 @@
+declare const require: (name: string) => any
+const { describe, test, expect } = require("bun:test")
+import type { BackgroundTask } from "./types"
+import { buildBackgroundTaskNotificationText } from "./background-task-notification-template"
+
+describe("notifyParentSession", () => {
+  test("displays INTERRUPTED for interrupted tasks", () => {
+    // given
+    const task: BackgroundTask = {
+      id: "test-task",
+      parentSessionID: "parent-session",
+      parentMessageID: "parent-message",
+      description: "Test task",
+      prompt: "Test prompt",
+      agent: "test-agent",
+      status: "interrupt",
+      startedAt: new Date(),
+      completedAt: new Date(),
+    }
+    const duration = "1s"
+    const statusText = task.status === "completed" ? "COMPLETED" : task.status === "interrupt" ? "INTERRUPTED" : "CANCELLED"
+    const allComplete = false
+    const remainingCount = 1
+    const completedTasks: BackgroundTask[] = []
+
+    // when
+    const notification = buildBackgroundTaskNotificationText({
+      task,
+      duration,
+      statusText,
+      allComplete,
+      remainingCount,
+      completedTasks,
+    })
+
+    // then
+    expect(notification).toContain("INTERRUPTED")
+  })
+})
--- a/src/features/background-agent/parent-session-notifier.ts
+++ b/src/features/background-agent/parent-session-notifier.ts
@@ -0,0 +1,103 @@
+import type { BackgroundTask } from "./types"
+import type { ResultHandlerContext } from "./result-handler-context"
+import { TASK_CLEANUP_DELAY_MS } from "./constants"
+import { createInternalAgentTextPart, log } from "../../shared"
+import { getTaskToastManager } from "../task-toast-manager"
+import { formatDuration } from "./duration-formatter"
+import { buildBackgroundTaskNotificationText } from "./background-task-notification-template"
+import { resolveParentSessionAgentAndModel } from "./parent-session-context-resolver"
+
+export async function notifyParentSession(
+  task: BackgroundTask,
+  ctx: ResultHandlerContext
+): Promise<void> {
+  const { client, state } = ctx
+
+  const duration = formatDuration(task.startedAt ?? task.completedAt ?? new Date(), task.completedAt)
+  log("[background-agent] notifyParentSession called for task:", task.id)
+
+  const toastManager = getTaskToastManager()
+  if (toastManager) {
+    toastManager.showCompletionToast({
+      id: task.id,
+      description: task.description,
+      duration,
+    })
+  }
+
+  const pendingSet = state.pendingByParent.get(task.parentSessionID)
+  if (pendingSet) {
+    pendingSet.delete(task.id)
+    if (pendingSet.size === 0) {
+      state.pendingByParent.delete(task.parentSessionID)
+    }
+  }
+
+  const allComplete = !pendingSet || pendingSet.size === 0
+  const remainingCount = pendingSet?.size ?? 0
+
+  const statusText = task.status === "completed" ? "COMPLETED" : task.status === "interrupt" ? "INTERRUPTED" : "CANCELLED"
+
+  const completedTasks = allComplete
+    ? Array.from(state.tasks.values()).filter(
+        (t) =>
+          t.parentSessionID === task.parentSessionID &&
+          t.status !== "running" &&
+          t.status !== "pending"
+      )
+    : []
+
+  const notification = buildBackgroundTaskNotificationText({
+    task,
+    duration,
+    statusText,
+    allComplete,
+    remainingCount,
+    completedTasks,
+  })
+
+  const { agent, model, tools } = await resolveParentSessionAgentAndModel({ client, task })
+
+  log("[background-agent] notifyParentSession context:", {
+    taskId: task.id,
+    resolvedAgent: agent,
+    resolvedModel: model,
+  })
+
+  try {
+    await client.session.promptAsync({
+      path: { id: task.parentSessionID },
+      body: {
+        noReply: !allComplete,
+        ...(agent !== undefined ? { agent } : {}),
+        ...(model !== undefined ? { model } : {}),
+        ...(tools ? { tools } : {}),
+        parts: [createInternalAgentTextPart(notification)],
+      },
+    })
+
+    log("[background-agent] Sent notification to parent session:", {
+      taskId: task.id,
+      allComplete,
+      noReply: !allComplete,
+    })
+  } catch (error) {
+    log("[background-agent] Failed to send notification:", error)
+  }
+
+  if (!allComplete) return
+
+  for (const completedTask of completedTasks) {
+    const taskId = completedTask.id
+    state.clearCompletionTimer(taskId)
+    const timer = setTimeout(() => {
+      state.completionTimers.delete(taskId)
+      if (state.tasks.has(taskId)) {
+        state.clearNotificationsForTask(taskId)
+        state.tasks.delete(taskId)
+        log("[background-agent] Removed completed task from memory:", taskId)
+      }
+    }, TASK_CLEANUP_DELAY_MS)
+    state.setCompletionTimer(taskId, timer)
+  }
+}
--- a/src/features/background-agent/result-handler-context.ts
+++ b/src/features/background-agent/result-handler-context.ts
@@ -0,0 +1,9 @@
+import type { OpencodeClient } from "./constants"
+import type { ConcurrencyManager } from "./concurrency"
+import type { TaskStateManager } from "./state"
+
+export interface ResultHandlerContext {
+  client: OpencodeClient
+  concurrencyManager: ConcurrencyManager
+  state: TaskStateManager
+}
--- a/src/features/background-agent/result-handler.ts
+++ b/src/features/background-agent/result-handler.ts
@@ -0,0 +1,7 @@
+export type { ResultHandlerContext } from "./result-handler-context"
+export { formatDuration } from "./duration-formatter"
+export { getMessageDir } from "../../shared"
+export { checkSessionTodos } from "./session-todo-checker"
+export { validateSessionHasOutput } from "./session-output-validator"
+export { tryCompleteTask } from "./background-task-completer"
+export { notifyParentSession } from "./parent-session-notifier"
--- a/src/features/background-agent/session-idle-event-handler.test.ts
+++ b/src/features/background-agent/session-idle-event-handler.test.ts
@@ -1,340 +0,0 @@
-import { describe, it, expect, mock } from "bun:test"
-
-import { handleSessionIdleBackgroundEvent } from "./session-idle-event-handler"
-import type { BackgroundTask } from "./types"
-import { MIN_IDLE_TIME_MS } from "./constants"
-
-function createRunningTask(overrides: Partial<BackgroundTask> = {}): BackgroundTask {
-  return {
-    id: "task-1",
-    sessionID: "ses-idle-1",
-    parentSessionID: "parent-ses-1",
-    parentMessageID: "msg-1",
-    description: "test idle handler",
-    prompt: "test",
-    agent: "explore",
-    status: "running",
-    startedAt: new Date(Date.now() - (MIN_IDLE_TIME_MS + 100)),
-    ...overrides,
-  }
-}
-
-describe("handleSessionIdleBackgroundEvent", () => {
-  describe("#given no sessionID in properties", () => {
-    it("#then should do nothing", () => {
-      //#given
-      const tryCompleteTask = mock(() => Promise.resolve(true))
-
-      //#when
-      handleSessionIdleBackgroundEvent({
-        properties: {},
-        findBySession: () => undefined,
-        idleDeferralTimers: new Map(),
-        validateSessionHasOutput: () => Promise.resolve(true),
-        checkSessionTodos: () => Promise.resolve(false),
-        tryCompleteTask,
-        emitIdleEvent: () => {},
-      })
-
-      //#then
-      expect(tryCompleteTask).not.toHaveBeenCalled()
-    })
-  })
-
-  describe("#given non-string sessionID in properties", () => {
-    it("#then should do nothing", () => {
-      //#given
-      const tryCompleteTask = mock(() => Promise.resolve(true))
-
-      //#when
-      handleSessionIdleBackgroundEvent({
-        properties: { sessionID: 123 },
-        findBySession: () => undefined,
-        idleDeferralTimers: new Map(),
-        validateSessionHasOutput: () => Promise.resolve(true),
-        checkSessionTodos: () => Promise.resolve(false),
-        tryCompleteTask,
-        emitIdleEvent: () => {},
-      })
-
-      //#then
-      expect(tryCompleteTask).not.toHaveBeenCalled()
-    })
-  })
-
-  describe("#given no task found for session", () => {
-    it("#then should do nothing", () => {
-      //#given
-      const tryCompleteTask = mock(() => Promise.resolve(true))
-
-      //#when
-      handleSessionIdleBackgroundEvent({
-        properties: { sessionID: "ses-unknown" },
-        findBySession: () => undefined,
-        idleDeferralTimers: new Map(),
-        validateSessionHasOutput: () => Promise.resolve(true),
-        checkSessionTodos: () => Promise.resolve(false),
-        tryCompleteTask,
-        emitIdleEvent: () => {},
-      })
-
-      //#then
-      expect(tryCompleteTask).not.toHaveBeenCalled()
-    })
-  })
-
-  describe("#given task is not running", () => {
-    it("#then should do nothing", () => {
-      //#given
-      const task = createRunningTask({ status: "completed" })
-      const tryCompleteTask = mock(() => Promise.resolve(true))
-
-      //#when
-      handleSessionIdleBackgroundEvent({
-        properties: { sessionID: task.sessionID! },
-        findBySession: () => task,
-        idleDeferralTimers: new Map(),
-        validateSessionHasOutput: () => Promise.resolve(true),
-        checkSessionTodos: () => Promise.resolve(false),
-        tryCompleteTask,
-        emitIdleEvent: () => {},
-      })
-
-      //#then
-      expect(tryCompleteTask).not.toHaveBeenCalled()
-    })
-  })
-
-  describe("#given task has no startedAt", () => {
-    it("#then should do nothing", () => {
-      //#given
-      const task = createRunningTask({ startedAt: undefined })
-      const tryCompleteTask = mock(() => Promise.resolve(true))
-
-      //#when
-      handleSessionIdleBackgroundEvent({
-        properties: { sessionID: task.sessionID! },
-        findBySession: () => task,
-        idleDeferralTimers: new Map(),
-        validateSessionHasOutput: () => Promise.resolve(true),
-        checkSessionTodos: () => Promise.resolve(false),
-        tryCompleteTask,
-        emitIdleEvent: () => {},
-      })
-
-      //#then
-      expect(tryCompleteTask).not.toHaveBeenCalled()
-    })
-  })
-
-  describe("#given elapsed time < MIN_IDLE_TIME_MS", () => {
-    it("#when idle fires early #then should defer with timer", () => {
-      //#given
-      const realDateNow = Date.now
-      const baseNow = realDateNow()
-      const task = createRunningTask({ startedAt: new Date(baseNow) })
-      const idleDeferralTimers = new Map<string, ReturnType<typeof setTimeout>>()
-      const emitIdleEvent = mock(() => {})
-
-      try {
-        Date.now = () => baseNow + (MIN_IDLE_TIME_MS - 100)
-
-        //#when
-        handleSessionIdleBackgroundEvent({
-          properties: { sessionID: task.sessionID! },
-          findBySession: () => task,
-          idleDeferralTimers,
-          validateSessionHasOutput: () => Promise.resolve(true),
-          checkSessionTodos: () => Promise.resolve(false),
-          tryCompleteTask: () => Promise.resolve(true),
-          emitIdleEvent,
-        })
-
-        //#then
-        expect(idleDeferralTimers.has(task.id)).toBe(true)
-        expect(emitIdleEvent).not.toHaveBeenCalled()
-      } finally {
-        clearTimeout(idleDeferralTimers.get(task.id)!)
-        Date.now = realDateNow
-      }
-    })
-
-    it("#when idle already deferred #then should not create duplicate timer", () => {
-      //#given
-      const realDateNow = Date.now
-      const baseNow = realDateNow()
-      const task = createRunningTask({ startedAt: new Date(baseNow) })
-      const existingTimer = setTimeout(() => {}, 99999)
-      const idleDeferralTimers = new Map<string, ReturnType<typeof setTimeout>>([
-        [task.id, existingTimer],
-      ])
-      const emitIdleEvent = mock(() => {})
-
-      try {
-        Date.now = () => baseNow + (MIN_IDLE_TIME_MS - 100)
-
-        //#when
-        handleSessionIdleBackgroundEvent({
-          properties: { sessionID: task.sessionID! },
-          findBySession: () => task,
-          idleDeferralTimers,
-          validateSessionHasOutput: () => Promise.resolve(true),
-          checkSessionTodos: () => Promise.resolve(false),
-          tryCompleteTask: () => Promise.resolve(true),
-          emitIdleEvent,
-        })
-
-        //#then
-        expect(idleDeferralTimers.get(task.id)).toBe(existingTimer)
-      } finally {
-        clearTimeout(existingTimer)
-        Date.now = realDateNow
-      }
-    })
-
-    it("#when deferred timer fires #then should emit idle event", async () => {
-      //#given
-      const realDateNow = Date.now
-      const baseNow = realDateNow()
-      const task = createRunningTask({ startedAt: new Date(baseNow) })
-      const idleDeferralTimers = new Map<string, ReturnType<typeof setTimeout>>()
-      const emitIdleEvent = mock(() => {})
-      const remainingMs = 50
-
-      try {
-        Date.now = () => baseNow + (MIN_IDLE_TIME_MS - remainingMs)
-
-        //#when
-        handleSessionIdleBackgroundEvent({
-          properties: { sessionID: task.sessionID! },
-          findBySession: () => task,
-          idleDeferralTimers,
-          validateSessionHasOutput: () => Promise.resolve(true),
-          checkSessionTodos: () => Promise.resolve(false),
-          tryCompleteTask: () => Promise.resolve(true),
-          emitIdleEvent,
-        })
-
-        //#then - wait for deferred timer
-        await new Promise((resolve) => setTimeout(resolve, remainingMs + 50))
-        expect(emitIdleEvent).toHaveBeenCalledWith(task.sessionID)
-        expect(idleDeferralTimers.has(task.id)).toBe(false)
-      } finally {
-        Date.now = realDateNow
-      }
-    })
-  })
-
-  describe("#given elapsed time >= MIN_IDLE_TIME_MS", () => {
-    it("#when session has valid output and no incomplete todos #then should complete task", async () => {
-      //#given
-      const task = createRunningTask()
-      const tryCompleteTask = mock(() => Promise.resolve(true))
-
-      //#when
-      handleSessionIdleBackgroundEvent({
-        properties: { sessionID: task.sessionID! },
-        findBySession: () => task,
-        idleDeferralTimers: new Map(),
-        validateSessionHasOutput: () => Promise.resolve(true),
-        checkSessionTodos: () => Promise.resolve(false),
-        tryCompleteTask,
-        emitIdleEvent: () => {},
-      })
-
-      //#then
-      await new Promise((resolve) => setTimeout(resolve, 10))
-      expect(tryCompleteTask).toHaveBeenCalledWith(task, "session.idle event")
-    })
-
-    it("#when session has no valid output #then should not complete task", async () => {
-      //#given
-      const task = createRunningTask()
-      const tryCompleteTask = mock(() => Promise.resolve(true))
-
-      //#when
-      handleSessionIdleBackgroundEvent({
-        properties: { sessionID: task.sessionID! },
-        findBySession: () => task,
-        idleDeferralTimers: new Map(),
-        validateSessionHasOutput: () => Promise.resolve(false),
-        checkSessionTodos: () => Promise.resolve(false),
-        tryCompleteTask,
-        emitIdleEvent: () => {},
-      })
-
-      //#then
-      await new Promise((resolve) => setTimeout(resolve, 10))
-      expect(tryCompleteTask).not.toHaveBeenCalled()
-    })
-
-    it("#when task has incomplete todos #then should not complete task", async () => {
-      //#given
-      const task = createRunningTask()
-      const tryCompleteTask = mock(() => Promise.resolve(true))
-
-      //#when
-      handleSessionIdleBackgroundEvent({
-        properties: { sessionID: task.sessionID! },
-        findBySession: () => task,
-        idleDeferralTimers: new Map(),
-        validateSessionHasOutput: () => Promise.resolve(true),
-        checkSessionTodos: () => Promise.resolve(true),
-        tryCompleteTask,
-        emitIdleEvent: () => {},
-      })
-
-      //#then
-      await new Promise((resolve) => setTimeout(resolve, 10))
-      expect(tryCompleteTask).not.toHaveBeenCalled()
-    })
-
-    it("#when task status changes during validation #then should not complete task", async () => {
-      //#given
-      const task = createRunningTask()
-      const tryCompleteTask = mock(() => Promise.resolve(true))
-
-      //#when
-      handleSessionIdleBackgroundEvent({
-        properties: { sessionID: task.sessionID! },
-        findBySession: () => task,
-        idleDeferralTimers: new Map(),
-        validateSessionHasOutput: async () => {
-          task.status = "completed"
-          return true
-        },
-        checkSessionTodos: () => Promise.resolve(false),
-        tryCompleteTask,
-        emitIdleEvent: () => {},
-      })
-
-      //#then
-      await new Promise((resolve) => setTimeout(resolve, 10))
-      expect(tryCompleteTask).not.toHaveBeenCalled()
-    })
-
-    it("#when task status changes during todo check #then should not complete task", async () => {
-      //#given
-      const task = createRunningTask()
-      const tryCompleteTask = mock(() => Promise.resolve(true))
-
-      //#when
-      handleSessionIdleBackgroundEvent({
-        properties: { sessionID: task.sessionID! },
-        findBySession: () => task,
-        idleDeferralTimers: new Map(),
-        validateSessionHasOutput: () => Promise.resolve(true),
-        checkSessionTodos: async () => {
-          task.status = "cancelled"
-          return false
-        },
-        tryCompleteTask,
-        emitIdleEvent: () => {},
-      })
-
-      //#then
-      await new Promise((resolve) => setTimeout(resolve, 10))
-      expect(tryCompleteTask).not.toHaveBeenCalled()
-    })
-  })
-})
--- a/src/features/background-agent/session-output-validator.ts
+++ b/src/features/background-agent/session-output-validator.ts
@@ -0,0 +1,89 @@
+import type { OpencodeClient } from "./constants"
+import { log } from "../../shared"
+
+type SessionMessagePart = {
+  type?: string
+  text?: string
+  content?: unknown
+}
+
+function isObject(value: unknown): value is Record<string, unknown> {
+  return typeof value === "object" && value !== null
+}
+
+function getMessageRole(message: unknown): string | undefined {
+  if (!isObject(message)) return undefined
+  const info = message["info"]
+  if (!isObject(info)) return undefined
+  const role = info["role"]
+  return typeof role === "string" ? role : undefined
+}
+
+function getMessageParts(message: unknown): SessionMessagePart[] {
+  if (!isObject(message)) return []
+  const parts = message["parts"]
+  if (!Array.isArray(parts)) return []
+
+  return parts
+    .filter((part): part is SessionMessagePart => isObject(part))
+    .map((part) => ({
+      type: typeof part["type"] === "string" ? part["type"] : undefined,
+      text: typeof part["text"] === "string" ? part["text"] : undefined,
+      content: part["content"],
+    }))
+}
+
+function partHasContent(part: SessionMessagePart): boolean {
+  if (part.type === "text" || part.type === "reasoning") {
+    return Boolean(part.text && part.text.trim().length > 0)
+  }
+  if (part.type === "tool") return true
+  if (part.type === "tool_result") {
+    if (typeof part.content === "string") return part.content.trim().length > 0
+    if (Array.isArray(part.content)) return part.content.length > 0
+    return Boolean(part.content)
+  }
+  return false
+}
+
+export async function validateSessionHasOutput(
+  client: OpencodeClient,
+  sessionID: string
+): Promise<boolean> {
+  try {
+    const response = await client.session.messages({
+      path: { id: sessionID },
+    })
+
+    const messagesRaw =
+      isObject(response) && "data" in response ? (response as { data?: unknown }).data : response
+    const messages = Array.isArray(messagesRaw) ? messagesRaw : []
+
+    const hasAssistantOrToolMessage = messages.some((message) => {
+      const role = getMessageRole(message)
+      return role === "assistant" || role === "tool"
+    })
+
+    if (!hasAssistantOrToolMessage) {
+      log("[background-agent] No assistant/tool messages found in session:", sessionID)
+      return false
+    }
+
+    const hasContent = messages.some((message) => {
+      const role = getMessageRole(message)
+      if (role !== "assistant" && role !== "tool") return false
+      const parts = getMessageParts(message)
+      return parts.some(partHasContent)
+    })
+
+    if (!hasContent) {
+      log("[background-agent] Messages exist but no content found in session:", sessionID)
+      return false
+    }
+
+    return true
+  } catch (error) {
+    log("[background-agent] Error validating session output:", error)
+    return true
+  }
+}
--- a/src/features/background-agent/session-task-cleanup.ts
+++ b/src/features/background-agent/session-task-cleanup.ts
@@ -0,0 +1,46 @@
+import { subagentSessions } from "../claude-code-session-state"
+import type { BackgroundTask } from "./types"
+
+export function cleanupTaskAfterSessionEnds(args: {
+  task: BackgroundTask
+  tasks: Map<string, BackgroundTask>
+  idleDeferralTimers: Map<string, ReturnType<typeof setTimeout>>
+  completionTimers: Map<string, ReturnType<typeof setTimeout>>
+  cleanupPendingByParent: (task: BackgroundTask) => void
+  clearNotificationsForTask: (taskId: string) => void
+  releaseConcurrencyKey?: (key: string) => void
+}): void {
+  const {
+    task,
+    tasks,
+    idleDeferralTimers,
+    completionTimers,
+    cleanupPendingByParent,
+    clearNotificationsForTask,
+    releaseConcurrencyKey,
+  } = args
+
+  const completionTimer = completionTimers.get(task.id)
+  if (completionTimer) {
+    clearTimeout(completionTimer)
+    completionTimers.delete(task.id)
+  }
+
+  const idleTimer = idleDeferralTimers.get(task.id)
+  if (idleTimer) {
+    clearTimeout(idleTimer)
+    idleDeferralTimers.delete(task.id)
+  }
+
+  if (task.concurrencyKey && releaseConcurrencyKey) {
+    releaseConcurrencyKey(task.concurrencyKey)
+    task.concurrencyKey = undefined
+  }
+
+  cleanupPendingByParent(task)
+  clearNotificationsForTask(task.id)
+  tasks.delete(task.id)
+  if (task.sessionID) {
+    subagentSessions.delete(task.sessionID)
+  }
+}
--- a/src/features/background-agent/session-todo-checker.ts
+++ b/src/features/background-agent/session-todo-checker.ts
@@ -0,0 +1,33 @@
+import type { OpencodeClient, Todo } from "./constants"
+
+function isTodo(value: unknown): value is Todo {
+  if (typeof value !== "object" || value === null) return false
+  const todo = value as Record<string, unknown>
+  return (
+    (typeof todo["id"] === "string" || todo["id"] === undefined) &&
+    typeof todo["content"] === "string" &&
+    typeof todo["status"] === "string" &&
+    typeof todo["priority"] === "string"
+  )
+}
+
+export async function checkSessionTodos(
+  client: OpencodeClient,
+  sessionID: string
+): Promise<boolean> {
+  try {
+    const response = await client.session.todo({
+      path: { id: sessionID },
+    })
+
+    const todosRaw = "data" in response ? response.data : response
+    if (!Array.isArray(todosRaw) || todosRaw.length === 0) return false
+
+    const incomplete = todosRaw
+      .filter(isTodo)
+      .filter((todo) => todo.status !== "completed" && todo.status !== "cancelled")
+    return incomplete.length > 0
+  } catch {
+    return false
+  }
+}
--- a/src/features/background-agent/spawner.ts
+++ b/src/features/background-agent/spawner.ts
@@ -61,7 +61,9 @@ export async function startTask(
  const createResult = await client.session.create({
    body: {
      parentID: input.parentSessionID,
-    } as Record<string, unknown>,
+      title: `Background: ${input.description}`,
+    // eslint-disable-next-line @typescript-eslint/no-explicit-any
+    } as any,
    query: {
      directory: parentDirectory,
    },
--- a/src/features/background-agent/spawner/background-session-creator.ts
+++ b/src/features/background-agent/spawner/background-session-creator.ts
@@ -0,0 +1,45 @@
+import type { OpencodeClient } from "../constants"
+import type { ConcurrencyManager } from "../concurrency"
+import type { LaunchInput } from "../types"
+import { log } from "../../../shared"
+
+export async function createBackgroundSession(options: {
+  client: OpencodeClient
+  input: LaunchInput
+  parentDirectory: string
+  concurrencyManager: ConcurrencyManager
+  concurrencyKey: string
+}): Promise<string> {
+  const { client, input, parentDirectory, concurrencyManager, concurrencyKey } = options
+
+  const body = {
+    parentID: input.parentSessionID,
+    title: `Background: ${input.description}`,
+  }
+
+  const createResult = await client.session
+    .create({
+      body,
+      query: {
+        directory: parentDirectory,
+      },
+    })
+    .catch((error: unknown) => {
+      concurrencyManager.release(concurrencyKey)
+      throw error
+    })
+
+  if (createResult.error) {
+    concurrencyManager.release(concurrencyKey)
+    throw new Error(`Failed to create background session: ${createResult.error}`)
+  }
+
+  if (!createResult.data?.id) {
+    concurrencyManager.release(concurrencyKey)
+    throw new Error("Failed to create background session: API returned no session ID")
+  }
+
+  const sessionID = createResult.data.id
+  log("[background-agent] Background session created", { sessionID })
+  return sessionID
+}
--- a/src/features/background-agent/spawner/concurrency-key-from-launch-input.ts
+++ b/src/features/background-agent/spawner/concurrency-key-from-launch-input.ts
@@ -0,0 +1,7 @@
+import type { LaunchInput } from "../types"
+
+export function getConcurrencyKeyFromLaunchInput(input: LaunchInput): string {
+  return input.model
+    ? `${input.model.providerID}/${input.model.modelID}`
+    : input.agent
+}
--- a/src/features/background-agent/spawner/spawner-context.ts
+++ b/src/features/background-agent/spawner/spawner-context.ts
@@ -0,0 +1,12 @@
+import type { BackgroundTask } from "../types"
+import type { ConcurrencyManager } from "../concurrency"
+import type { OpencodeClient, OnSubagentSessionCreated } from "../constants"
+
+export interface SpawnerContext {
+  client: OpencodeClient
+  directory: string
+  concurrencyManager: ConcurrencyManager
+  tmuxEnabled: boolean
+  onSubagentSessionCreated?: OnSubagentSessionCreated
+  onTaskError: (task: BackgroundTask, error: Error) => void
+}
--- a/src/features/background-agent/spawner/tmux-callback-invoker.ts
+++ b/src/features/background-agent/spawner/tmux-callback-invoker.ts
@@ -0,0 +1,40 @@
+import { setTimeout } from "timers/promises"
+import type { OnSubagentSessionCreated } from "../constants"
+import { TMUX_CALLBACK_DELAY_MS } from "../constants"
+import { log } from "../../../shared"
+import { isInsideTmux } from "../../../shared/tmux"
+
+export async function maybeInvokeTmuxCallback(options: {
+  onSubagentSessionCreated?: OnSubagentSessionCreated
+  tmuxEnabled: boolean
+  sessionID: string
+  parentID: string
+  title: string
+}): Promise<void> {
+  const { onSubagentSessionCreated, tmuxEnabled, sessionID, parentID, title } = options
+
+  log("[background-agent] tmux callback check", {
+    hasCallback: !!onSubagentSessionCreated,
+    tmuxEnabled,
+    isInsideTmux: isInsideTmux(),
+    sessionID,
+    parentID,
+  })
+
+  if (!onSubagentSessionCreated || !tmuxEnabled || !isInsideTmux()) {
+    log("[background-agent] SKIP tmux callback - conditions not met")
+    return
+  }
+
+  log("[background-agent] Invoking tmux callback NOW", { sessionID })
+  await onSubagentSessionCreated({
+    sessionID,
+    parentID,
+    title,
+  }).catch((error: unknown) => {
+    log("[background-agent] Failed to spawn tmux pane:", error)
+  })
+
+  log("[background-agent] tmux callback completed, waiting")
+  await setTimeout(TMUX_CALLBACK_DELAY_MS)
+}
--- a/src/features/boulder-state/storage.test.ts
+++ b/src/features/boulder-state/storage.test.ts
@@ -269,71 +269,6 @@ describe("boulder-state", () => {
      expect(progress.isComplete).toBe(false)
    })

-    test("should count space-indented unchecked checkbox", () => {
-      // given - plan file with a two-space indented checkbox
-      const planPath = join(TEST_DIR, "space-indented-plan.md")
-      writeFileSync(planPath, `# Plan
-  - [ ] indented task
-`)
-
-      // when
-      const progress = getPlanProgress(planPath)
-
-      // then
-      expect(progress.total).toBe(1)
-      expect(progress.completed).toBe(0)
-      expect(progress.isComplete).toBe(false)
-    })
-
-    test("should count tab-indented unchecked checkbox", () => {
-      // given - plan file with a tab-indented checkbox
-      const planPath = join(TEST_DIR, "tab-indented-plan.md")
-      writeFileSync(planPath, `# Plan
-	- [ ] tab-indented task
-`)
-
-      // when
-      const progress = getPlanProgress(planPath)
-
-      // then
-      expect(progress.total).toBe(1)
-      expect(progress.completed).toBe(0)
-      expect(progress.isComplete).toBe(false)
-    })
-
-    test("should count mixed top-level checked and indented unchecked checkboxes", () => {
-      // given - plan file with checked top-level and unchecked indented task
-      const planPath = join(TEST_DIR, "mixed-indented-plan.md")
-      writeFileSync(planPath, `# Plan
- [x] top-level completed task
-  - [ ] nested unchecked task
-`)
-
-      // when
-      const progress = getPlanProgress(planPath)
-
-      // then
-      expect(progress.total).toBe(2)
-      expect(progress.completed).toBe(1)
-      expect(progress.isComplete).toBe(false)
-    })
-
-    test("should count space-indented completed checkbox", () => {
-      // given - plan file with a two-space indented completed checkbox
-      const planPath = join(TEST_DIR, "indented-completed-plan.md")
-      writeFileSync(planPath, `# Plan
-  - [x] indented completed task
-`)
-
-      // when
-      const progress = getPlanProgress(planPath)
-
-      // then
-      expect(progress.total).toBe(1)
-      expect(progress.completed).toBe(1)
-      expect(progress.isComplete).toBe(true)
-    })
-
    test("should return isComplete true when all checked", () => {
      // given - all tasks completed
      const planPath = join(TEST_DIR, "complete-plan.md")
--- a/src/features/boulder-state/storage.ts
+++ b/src/features/boulder-state/storage.ts
@@ -121,8 +121,8 @@ export function getPlanProgress(planPath: string): PlanProgress {
    const content = readFileSync(planPath, "utf-8")
    
    // Match markdown checkboxes: - [ ] or - [x] or - [X]
-    const uncheckedMatches = content.match(/^\s*[-*]\s*\[\s*\]/gm) || []
-    const checkedMatches = content.match(/^\s*[-*]\s*\[[xX]\]/gm) || []
+    const uncheckedMatches = content.match(/^[-*]\s*\[\s*\]/gm) || []
+    const checkedMatches = content.match(/^[-*]\s*\[[xX]\]/gm) || []

    const total = uncheckedMatches.length + checkedMatches.length
    const completed = checkedMatches.length
@@ -150,8 +150,7 @@ export function getPlanName(planPath: string): string {
 export function createBoulderState(
  planPath: string,
  sessionId: string,
-  agent?: string,
-  worktreePath?: string,
+  agent?: string
 ): BoulderState {
  return {
    active_plan: planPath,
@@ -159,6 +158,5 @@ export function createBoulderState(
    session_ids: [sessionId],
    plan_name: getPlanName(planPath),
    ...(agent !== undefined ? { agent } : {}),
-    ...(worktreePath !== undefined ? { worktree_path: worktreePath } : {}),
  }
 }
--- a/src/features/boulder-state/types.ts
+++ b/src/features/boulder-state/types.ts
@@ -16,8 +16,6 @@ export interface BoulderState {
  plan_name: string
  /** Agent type to use when resuming (e.g., 'atlas') */
  agent?: string
-  /** Absolute path to the git worktree root where work happens */
-  worktree_path?: string
 }

 export interface PlanProgress {
--- a/src/features/builtin-commands/templates/start-work.ts
+++ b/src/features/builtin-commands/templates/start-work.ts
@@ -1,14 +1,5 @@
 export const START_WORK_TEMPLATE = `You are starting a Sisyphus work session.

-## ARGUMENTS
-
- \`/start-work [plan-name] [--worktree <path>]\`
-  - \`plan-name\` (optional): name or partial match of the plan to start
-  - \`--worktree <path>\` (optional): absolute path to an existing git worktree to work in
-    - If specified and valid: hook pre-sets worktree_path in boulder.json
-    - If specified but invalid: you must run \`git worktree add <path> <branch>\` first
-    - If omitted: you MUST choose or create a worktree (see Worktree Setup below)
-
 ## WHAT TO DO

 1. **Find available plans**: Search for Prometheus-generated plan files at \`.sisyphus/plans/\`
@@ -24,24 +15,17 @@ export const START_WORK_TEMPLATE = `You are starting a Sisyphus work session.
     - If ONE plan: auto-select it
     - If MULTIPLE plans: show list with timestamps, ask user to select

-4. **Worktree Setup** (when \`worktree_path\` not already set in boulder.json):
-   1. \`git worktree list --porcelain\` — see available worktrees
-   2. Create: \`git worktree add <absolute-path> <branch-or-HEAD>\`
-   3. Update boulder.json to add \`"worktree_path": "<absolute-path>"\`
-   4. All work happens inside that worktree directory
-
-5. **Create/Update boulder.json**:
+4. **Create/Update boulder.json**:
   \`\`\`json
   {
     "active_plan": "/absolute/path/to/plan.md",
     "started_at": "ISO_TIMESTAMP",
     "session_ids": ["session_id_1", "session_id_2"],
-     "plan_name": "plan-name",
-     "worktree_path": "/absolute/path/to/git/worktree"
+     "plan_name": "plan-name"
   }
   \`\`\`

-6. **Read the plan file** and start executing tasks according to atlas workflow
+5. **Read the plan file** and start executing tasks according to atlas workflow

 ## OUTPUT FORMAT

@@ -65,7 +49,6 @@ Resuming Work Session
 Active Plan: {plan-name}
 Progress: {completed}/{total} tasks
 Sessions: {count} (appending current session)
-Worktree: {worktree_path}

 Reading plan and continuing from last incomplete task...
 \`\`\`
@@ -77,7 +60,6 @@ Starting Work Session
 Plan: {plan-name}
 Session ID: {session_id}
 Started: {timestamp}
-Worktree: {worktree_path}

 Reading plan and beginning execution...
 \`\`\`
@@ -86,6 +68,5 @@ Reading plan and beginning execution...

 - The session_id is injected by the hook - use it directly
 - Always update boulder.json BEFORE starting work
- Always set worktree_path in boulder.json before executing any tasks
 - Read the FULL plan file before delegating any tasks
 - Follow atlas delegation protocols (7-section format)`
--- a/src/features/claude-tasks/AGENTS.md
+++ b/src/features/claude-tasks/AGENTS.md
@@ -1,6 +1,6 @@
 # src/features/claude-tasks/ — Task Schema + Storage

-**Generated:** 2026-02-24
+**Generated:** 2026-02-21

 ## OVERVIEW

--- a/src/features/claude-tasks/index.ts
+++ b/src/features/claude-tasks/index.ts
@@ -0,0 +1,3 @@
+export * from "./types"
+export * from "./storage"
+export * from "./session-storage"
--- a/src/features/mcp-oauth/AGENTS.md
+++ b/src/features/mcp-oauth/AGENTS.md
@@ -1,6 +1,6 @@
 # src/features/mcp-oauth/ — OAuth 2.0 + PKCE + DCR for MCP Servers

-**Generated:** 2026-02-24
+**Generated:** 2026-02-21

 ## OVERVIEW

--- a/src/features/mcp-oauth/index.ts
+++ b/src/features/mcp-oauth/index.ts
@@ -0,0 +1,3 @@
+export * from "./schema"
+export * from "./oauth-authorization-flow"
+export * from "./provider"
--- a/src/features/opencode-skill-loader/AGENTS.md
+++ b/src/features/opencode-skill-loader/AGENTS.md
@@ -1,6 +1,6 @@
 # src/features/opencode-skill-loader/ — 4-Scope Skill Discovery

-**Generated:** 2026-02-24
+**Generated:** 2026-02-21

 ## OVERVIEW

--- a/src/features/task-toast-manager/manager.ts
+++ b/src/features/task-toast-manager/manager.ts
@@ -4,12 +4,6 @@ import type { ConcurrencyManager } from "../background-agent/concurrency"

 type OpencodeClient = PluginInput["client"]

-type ClientWithTui = {
-  tui?: {
-    showToast: (opts: { body: { title: string; message: string; variant: string; duration: number } }) => Promise<unknown>
-  }
-}
-
 export class TaskToastManager {
  private tasks: Map<string, TrackedTask> = new Map()
  private client: OpencodeClient
@@ -176,7 +170,8 @@ export class TaskToastManager {
   * Show consolidated toast with all running/queued tasks
   */
  private showTaskListToast(newTask: TrackedTask): void {
-    const tuiClient = this.client as ClientWithTui
+    // eslint-disable-next-line @typescript-eslint/no-explicit-any
+    const tuiClient = this.client as any
    if (!tuiClient.tui?.showToast) return

    const message = this.buildTaskListMessage(newTask)
@@ -201,7 +196,8 @@ export class TaskToastManager {
   * Show task completion toast
   */
  showCompletionToast(task: { id: string; description: string; duration: string }): void {
-    const tuiClient = this.client as ClientWithTui
+    // eslint-disable-next-line @typescript-eslint/no-explicit-any
+    const tuiClient = this.client as any
    if (!tuiClient.tui?.showToast) return

    this.removeTask(task.id)
--- a/src/features/tmux-subagent/AGENTS.md
+++ b/src/features/tmux-subagent/AGENTS.md
@@ -1,6 +1,6 @@
 # src/features/tmux-subagent/ — Tmux Pane Management

-**Generated:** 2026-02-24
+**Generated:** 2026-02-21

 ## OVERVIEW

--- a/src/hooks/AGENTS.md
+++ b/src/hooks/AGENTS.md
@@ -1,14 +1,14 @@
-# src/hooks/ — 46 Lifecycle Hooks
+# src/hooks/ — 44 Lifecycle Hooks

-**Generated:** 2026-02-24
+**Generated:** 2026-02-21

 ## OVERVIEW

-46 hooks across 39 directories + 6 standalone files. Three-tier composition: Core(37) + Continuation(7) + Skill(2). All hooks follow `createXXXHook(deps) → HookFunction` factory pattern.
+44 hooks across 39 directories + 6 standalone files. Three-tier composition: Core(35) + Continuation(7) + Skill(2). All hooks follow `createXXXHook(deps) → HookFunction` factory pattern.

 ## HOOK TIERS

-### Tier 1: Session Hooks (23) — `create-session-hooks.ts`
+### Tier 1: Session Hooks (22) — `create-session-hooks.ts`
 ## STRUCTURE
 ```
 hooks/
@@ -70,12 +70,11 @@ hooks/
 | questionLabelTruncator | tool.execute.before | Truncate long question labels |
 | taskResumeInfo | chat.message | Inject task context on resume |
 | anthropicEffort | chat.params | Adjust reasoning effort level |
-| modelFallback | chat.params | Provider-level model fallback on errors |
-| noSisyphusGpt | chat.message | Block Sisyphus from using GPT models (toast warning) |
-| noHephaestusNonGpt | chat.message | Block Hephaestus from using non-GPT models |
-| runtimeFallback | event | Auto-switch models on API provider errors |
+| jsonErrorRecovery | tool.execute.after | Detect JSON parse errors, inject correction reminder |
+| sisyphusGptHephaestusReminder | chat.message | Toast warning when Sisyphus uses GPT model |
+| taskReminder | tool.execute.after | Remind about task tools after 10 turns without usage |

-### Tier 2: Tool Guard Hooks (10) — `create-tool-guard-hooks.ts`
+### Tier 2: Tool Guard Hooks (9) — `create-tool-guard-hooks.ts`

 | Hook | Event | Purpose |
 |------|-------|---------|
@@ -88,7 +87,6 @@ hooks/
 | tasksTodowriteDisabler | tool.execute.before | Disable TodoWrite when task system active |
 | writeExistingFileGuard | tool.execute.before | Require Read before Write on existing files |
 | hashlineReadEnhancer | tool.execute.after | Enhance Read output with line hashes |
-| jsonErrorRecovery | tool.execute.after | Detect JSON parse errors, inject correction reminder |

 ### Tier 3: Transform Hooks (4) — `create-transform-hooks.ts`

--- a/src/hooks/anthropic-context-window-limit-recovery/AGENTS.md
+++ b/src/hooks/anthropic-context-window-limit-recovery/AGENTS.md
@@ -1,6 +1,6 @@
 # src/hooks/anthropic-context-window-limit-recovery/ — Multi-Strategy Context Recovery

-**Generated:** 2026-02-24
+**Generated:** 2026-02-21

 ## OVERVIEW

--- a/src/hooks/anthropic-context-window-limit-recovery/executor.ts
+++ b/src/hooks/anthropic-context-window-limit-recovery/executor.ts
@@ -1,5 +1,4 @@
 import type { AutoCompactState } from "./types";
-import type { OhMyOpenCodeConfig } from "../../config";
 import type { ExperimentalConfig } from "../../config";
 import { TRUNCATE_CONFIG } from "./types";

@@ -16,15 +15,15 @@ export async function executeCompact(
  sessionID: string,
  msg: Record<string, unknown>,
  autoCompactState: AutoCompactState,
-  client: Client,
+  // eslint-disable-next-line @typescript-eslint/no-explicit-any
+  client: any,
  directory: string,
-  pluginConfig: OhMyOpenCodeConfig,
-  _experimental?: ExperimentalConfig
+  experimental?: ExperimentalConfig,
 ): Promise<void> {
-  void _experimental
+  void experimental

  if (autoCompactState.compactionInProgress.has(sessionID)) {
-    await client.tui
+    await (client as Client).tui
      .showToast({
        body: {
          title: "Compact In Progress",
@@ -56,7 +55,7 @@ export async function executeCompact(
      const result = await runAggressiveTruncationStrategy({
        sessionID,
        autoCompactState,
-        client: client,
+        client: client as Client,
        directory,
        truncateAttempt: truncateState.truncateAttempt,
        currentTokens: errorData.currentTokens,
@@ -71,9 +70,8 @@ export async function executeCompact(
      sessionID,
      msg,
      autoCompactState,
-      client: client,
+      client: client as Client,
      directory,
-      pluginConfig,
      errorType: errorData?.errorType,
      messageIndex: errorData?.messageIndex,
    })
--- a/src/hooks/anthropic-context-window-limit-recovery/recovery-hook.ts
+++ b/src/hooks/anthropic-context-window-limit-recovery/recovery-hook.ts
@@ -1,7 +1,6 @@
 import type { PluginInput } from "@opencode-ai/plugin"
-import type { Client } from "./client"
 import type { AutoCompactState, ParsedTokenLimitError } from "./types"
-import type { ExperimentalConfig, OhMyOpenCodeConfig } from "../../config"
+import type { ExperimentalConfig } from "../../config"
 import { parseAnthropicTokenLimitError } from "./parser"
 import { executeCompact, getLastAssistant } from "./executor"
 import { attemptDeduplicationRecovery } from "./deduplication-recovery"
@@ -9,7 +8,6 @@ import { log } from "../../shared/logger"

 export interface AnthropicContextWindowLimitRecoveryOptions {
  experimental?: ExperimentalConfig
-  pluginConfig: OhMyOpenCodeConfig
 }

 function createRecoveryState(): AutoCompactState {
@@ -30,7 +28,6 @@ export function createAnthropicContextWindowLimitRecoveryHook(
 ) {
  const autoCompactState = createRecoveryState()
  const experimental = options?.experimental
-  const pluginConfig = options?.pluginConfig!
  const pendingCompactionTimeoutBySession = new Map<string, ReturnType<typeof setTimeout>>()

  const eventHandler = async ({ event }: { event: { type: string; properties?: unknown } }) => {
@@ -92,9 +89,8 @@ export function createAnthropicContextWindowLimitRecoveryHook(
            sessionID,
            { providerID, modelID },
            autoCompactState,
-            ctx.client as Client,
+            ctx.client,
            ctx.directory,
-            pluginConfig,
            experimental,
          )
        }, 300)
@@ -160,9 +156,8 @@ export function createAnthropicContextWindowLimitRecoveryHook(
        sessionID,
        { providerID, modelID },
        autoCompactState,
-        ctx.client as Client,
+        ctx.client,
        ctx.directory,
-        pluginConfig,
        experimental,
      )
    }
--- a/src/hooks/anthropic-context-window-limit-recovery/state.ts
+++ b/src/hooks/anthropic-context-window-limit-recovery/state.ts
@@ -6,7 +6,7 @@ export function getOrCreateRetryState(
 ): RetryState {
  let state = autoCompactState.retryStateBySession.get(sessionID)
  if (!state) {
-    state = { attempt: 0, lastAttemptTime: 0, firstAttemptTime: 0 }
+    state = { attempt: 0, lastAttemptTime: 0 }
    autoCompactState.retryStateBySession.set(sessionID, state)
  }
  return state
--- a/src/hooks/anthropic-context-window-limit-recovery/summarize-retry-strategy.test.ts
+++ b/src/hooks/anthropic-context-window-limit-recovery/summarize-retry-strategy.test.ts
@@ -1,122 +0,0 @@
-import { afterEach, beforeEach, describe, expect, mock, test } from "bun:test"
-import { runSummarizeRetryStrategy } from "./summarize-retry-strategy"
-import type { AutoCompactState, ParsedTokenLimitError, RetryState } from "./types"
-import type { OhMyOpenCodeConfig } from "../../config"
-
-type TimeoutCall = {
-  delay: number
-}
-
-function createAutoCompactState(): AutoCompactState {
-  return {
-    pendingCompact: new Set<string>(),
-    errorDataBySession: new Map<string, ParsedTokenLimitError>(),
-    retryStateBySession: new Map<string, RetryState>(),
-    truncateStateBySession: new Map(),
-    emptyContentAttemptBySession: new Map(),
-    compactionInProgress: new Set<string>(),
-  }
-}
-
-describe("runSummarizeRetryStrategy", () => {
-  const sessionID = "ses_retry_timeout"
-  const directory = "/tmp"
-  let autoCompactState: AutoCompactState
-
-  const summarizeMock = mock(() => Promise.resolve())
-  const showToastMock = mock(() => Promise.resolve())
-  const client = {
-    session: {
-      summarize: summarizeMock,
-      messages: mock(() => Promise.resolve({ data: [] })),
-      promptAsync: mock(() => Promise.resolve()),
-      revert: mock(() => Promise.resolve()),
-    },
-    tui: {
-      showToast: showToastMock,
-    },
-  }
-
-  beforeEach(() => {
-    autoCompactState = createAutoCompactState()
-    summarizeMock.mockReset()
-    showToastMock.mockReset()
-    summarizeMock.mockResolvedValue(undefined)
-    showToastMock.mockResolvedValue(undefined)
-  })
-
-  afterEach(() => {
-    globalThis.setTimeout = originalSetTimeout
-  })
-
-  const originalSetTimeout = globalThis.setTimeout
-
-  test("stops retries when total summarize timeout is exceeded", async () => {
-    //#given
-    autoCompactState.pendingCompact.add(sessionID)
-    autoCompactState.errorDataBySession.set(sessionID, {
-      currentTokens: 250000,
-      maxTokens: 200000,
-      errorType: "token_limit_exceeded",
-    })
-    autoCompactState.retryStateBySession.set(sessionID, {
-      attempt: 1,
-      lastAttemptTime: Date.now(),
-      firstAttemptTime: Date.now() - 130000,
-    })
-
-    //#when
-    await runSummarizeRetryStrategy({
-      sessionID,
-      msg: { providerID: "anthropic", modelID: "claude-sonnet-4-6" },
-      autoCompactState,
-      client: client as never,
-      directory,
-      pluginConfig: {} as OhMyOpenCodeConfig,
-    })
-
-    //#then
-    expect(summarizeMock).not.toHaveBeenCalled()
-    expect(autoCompactState.pendingCompact.has(sessionID)).toBe(false)
-    expect(autoCompactState.errorDataBySession.has(sessionID)).toBe(false)
-    expect(autoCompactState.retryStateBySession.has(sessionID)).toBe(false)
-    expect(showToastMock).toHaveBeenCalledWith(
-      expect.objectContaining({
-        body: expect.objectContaining({
-          title: "Auto Compact Timed Out",
-        }),
-      }),
-    )
-  })
-
-  test("caps retry delay by remaining total timeout window", async () => {
-    //#given
-    const timeoutCalls: TimeoutCall[] = []
-    globalThis.setTimeout = ((_: (...args: unknown[]) => void, delay?: number) => {
-      timeoutCalls.push({ delay: delay ?? 0 })
-      return 1 as unknown as ReturnType<typeof setTimeout>
-    }) as typeof setTimeout
-
-    autoCompactState.retryStateBySession.set(sessionID, {
-      attempt: 1,
-      lastAttemptTime: Date.now(),
-      firstAttemptTime: Date.now() - 119700,
-    })
-    summarizeMock.mockRejectedValueOnce(new Error("rate limited"))
-
-    //#when
-    await runSummarizeRetryStrategy({
-      sessionID,
-      msg: { providerID: "anthropic", modelID: "claude-sonnet-4-6" },
-      autoCompactState,
-      client: client as never,
-      directory,
-      pluginConfig: {} as OhMyOpenCodeConfig,
-    })
-
-    //#then
-    expect(timeoutCalls.length).toBe(1)
-    expect(timeoutCalls[0]!.delay).toBeGreaterThan(0)
-    expect(timeoutCalls[0]!.delay).toBeLessThanOrEqual(500)
-  })
-})
--- a/src/hooks/anthropic-context-window-limit-recovery/summarize-retry-strategy.ts
+++ b/src/hooks/anthropic-context-window-limit-recovery/summarize-retry-strategy.ts
@@ -1,46 +1,20 @@
 import type { AutoCompactState } from "./types"
-import type { OhMyOpenCodeConfig } from "../../config"
 import { RETRY_CONFIG } from "./types"
 import type { Client } from "./client"
 import { clearSessionState, getEmptyContentAttempt, getOrCreateRetryState } from "./state"
 import { sanitizeEmptyMessagesBeforeSummarize } from "./message-builder"
 import { fixEmptyMessages } from "./empty-content-recovery"

-import { resolveCompactionModel } from "../shared/compaction-model-resolver"
-
-const SUMMARIZE_RETRY_TOTAL_TIMEOUT_MS = 120_000
 export async function runSummarizeRetryStrategy(params: {
  sessionID: string
  msg: Record<string, unknown>
  autoCompactState: AutoCompactState
  client: Client
  directory: string
-  pluginConfig: OhMyOpenCodeConfig
  errorType?: string
  messageIndex?: number
 }): Promise<void> {
  const retryState = getOrCreateRetryState(params.autoCompactState, params.sessionID)
-  const now = Date.now()
-
-  if (retryState.firstAttemptTime === 0) {
-    retryState.firstAttemptTime = now
-  }
-
-  const elapsedTimeMs = now - retryState.firstAttemptTime
-  if (elapsedTimeMs >= SUMMARIZE_RETRY_TOTAL_TIMEOUT_MS) {
-    clearSessionState(params.autoCompactState, params.sessionID)
-    await params.client.tui
-      .showToast({
-        body: {
-          title: "Auto Compact Timed Out",
-          message: "Compaction retries exceeded the timeout window. Please start a new session.",
-          variant: "error",
-          duration: 5000,
-        },
-      })
-      .catch(() => {})
-    return
-  }

  if (params.errorType?.includes("non-empty content")) {
    const attempt = getEmptyContentAttempt(params.autoCompactState, params.sessionID)
@@ -75,7 +49,6 @@ export async function runSummarizeRetryStrategy(params: {

  if (Date.now() - retryState.lastAttemptTime > 300000) {
    retryState.attempt = 0
-    retryState.firstAttemptTime = Date.now()
    params.autoCompactState.truncateStateBySession.delete(params.sessionID)
  }

@@ -101,14 +74,7 @@ export async function runSummarizeRetryStrategy(params: {
          })
          .catch(() => {})

-        const { providerID: targetProviderID, modelID: targetModelID } = resolveCompactionModel(
-          params.pluginConfig,
-          params.sessionID,
-          providerID,
-          modelID
-        )
-
-        const summarizeBody = { providerID: targetProviderID, modelID: targetModelID, auto: true }
+        const summarizeBody = { providerID, modelID, auto: true }
        await params.client.session.summarize({
          path: { id: params.sessionID },
          body: summarizeBody as never,
@@ -116,26 +82,10 @@ export async function runSummarizeRetryStrategy(params: {
        })
        return
      } catch {
-        const remainingTimeMs = SUMMARIZE_RETRY_TOTAL_TIMEOUT_MS - (Date.now() - retryState.firstAttemptTime)
-        if (remainingTimeMs <= 0) {
-          clearSessionState(params.autoCompactState, params.sessionID)
-          await params.client.tui
-            .showToast({
-              body: {
-                title: "Auto Compact Timed Out",
-                message: "Compaction retries exceeded the timeout window. Please start a new session.",
-                variant: "error",
-                duration: 5000,
-              },
-            })
-            .catch(() => {})
-          return
-        }
-
        const delay =
          RETRY_CONFIG.initialDelayMs *
          Math.pow(RETRY_CONFIG.backoffFactor, retryState.attempt - 1)
-        const cappedDelay = Math.min(delay, RETRY_CONFIG.maxDelayMs, remainingTimeMs)
+        const cappedDelay = Math.min(delay, RETRY_CONFIG.maxDelayMs)

        setTimeout(() => {
          void runSummarizeRetryStrategy(params)
--- a/src/hooks/anthropic-context-window-limit-recovery/types.ts
+++ b/src/hooks/anthropic-context-window-limit-recovery/types.ts
@@ -11,7 +11,6 @@ export interface ParsedTokenLimitError {
 export interface RetryState {
  attempt: number
  lastAttemptTime: number
-  firstAttemptTime: number
 }

 export interface TruncateState {
--- a/src/hooks/atlas/AGENTS.md
+++ b/src/hooks/atlas/AGENTS.md
@@ -1,6 +1,6 @@
 # src/hooks/atlas/ — Master Boulder Orchestrator

-**Generated:** 2026-02-24
+**Generated:** 2026-02-21

 ## OVERVIEW

--- a/src/hooks/atlas/boulder-continuation-injector.ts
+++ b/src/hooks/atlas/boulder-continuation-injector.ts
@@ -14,7 +14,6 @@ export async function injectBoulderContinuation(input: {
  remaining: number
  total: number
  agent?: string
-  worktreePath?: string
  backgroundManager?: BackgroundManager
  sessionState: SessionState
 }): Promise<void> {
@@ -25,7 +24,6 @@ export async function injectBoulderContinuation(input: {
    remaining,
    total,
    agent,
-    worktreePath,
    backgroundManager,
    sessionState,
  } = input
@@ -39,11 +37,9 @@ export async function injectBoulderContinuation(input: {
    return
  }

-  const worktreeContext = worktreePath ? `\n\n[Worktree: ${worktreePath}]` : ""
  const prompt =
    BOULDER_CONTINUATION_PROMPT.replace(/{PLAN_NAME}/g, planName) +
-    `\n\n[Status: ${total - remaining}/${total} completed, ${remaining} remaining]` +
-    worktreeContext
+    `\n\n[Status: ${total - remaining}/${total} completed, ${remaining} remaining]`

  try {
    log(`[${HOOK_NAME}] Injecting boulder continuation`, { sessionID, planName, remaining })
@@ -66,7 +62,6 @@ export async function injectBoulderContinuation(input: {
    log(`[${HOOK_NAME}] Boulder continuation injected`, { sessionID })
  } catch (err) {
    sessionState.promptFailureCount += 1
-    sessionState.lastFailureAt = Date.now()
    log(`[${HOOK_NAME}] Boulder continuation failed`, {
      sessionID,
      error: String(err),
--- a/src/hooks/atlas/event-handler.ts
+++ b/src/hooks/atlas/event-handler.ts
@@ -10,7 +10,6 @@ import { getLastAgentFromSession } from "./session-last-agent"
 import type { AtlasHookOptions, SessionState } from "./types"

 const CONTINUATION_COOLDOWN_MS = 5000
-const FAILURE_BACKOFF_MS = 5 * 60 * 1000

 export function createAtlasEventHandler(input: {
  ctx: PluginInput
@@ -54,7 +53,6 @@ export function createAtlasEventHandler(input: {
      }

      const state = getState(sessionID)
-      const now = Date.now()

      if (state.lastEventWasAbortError) {
        state.lastEventWasAbortError = false
@@ -63,18 +61,11 @@ export function createAtlasEventHandler(input: {
      }

      if (state.promptFailureCount >= 2) {
-        const timeSinceLastFailure = state.lastFailureAt !== undefined ? now - state.lastFailureAt : Number.POSITIVE_INFINITY
-        if (timeSinceLastFailure < FAILURE_BACKOFF_MS) {
-          log(`[${HOOK_NAME}] Skipped: continuation in backoff after repeated failures`, {
-            sessionID,
-            promptFailureCount: state.promptFailureCount,
-            backoffRemaining: FAILURE_BACKOFF_MS - timeSinceLastFailure,
-          })
-          return
-        }
-
-        state.promptFailureCount = 0
-        state.lastFailureAt = undefined
+        log(`[${HOOK_NAME}] Skipped: continuation disabled after repeated prompt failures`, {
+          sessionID,
+          promptFailureCount: state.promptFailureCount,
+        })
+        return
      }

      const backgroundManager = options?.backgroundManager
@@ -101,15 +92,17 @@ export function createAtlasEventHandler(input: {
      const lastAgentKey = getAgentConfigKey(lastAgent ?? "")
      const requiredAgent = getAgentConfigKey(boulderState.agent ?? "atlas")
      const lastAgentMatchesRequired = lastAgentKey === requiredAgent
+      const boulderAgentWasNotExplicitlySet = boulderState.agent === undefined
      const boulderAgentDefaultsToAtlas = requiredAgent === "atlas"
      const lastAgentIsSisyphus = lastAgentKey === "sisyphus"
-      const allowSisyphusForAtlasBoulder = boulderAgentDefaultsToAtlas && lastAgentIsSisyphus
-      const agentMatches = lastAgentMatchesRequired || allowSisyphusForAtlasBoulder
+      const allowSisyphusWhenDefaultAtlas = boulderAgentWasNotExplicitlySet && boulderAgentDefaultsToAtlas && lastAgentIsSisyphus
+      const agentMatches = lastAgentMatchesRequired || allowSisyphusWhenDefaultAtlas
      if (!agentMatches) {
        log(`[${HOOK_NAME}] Skipped: last agent does not match boulder agent`, {
          sessionID,
          lastAgent: lastAgent ?? "unknown",
          requiredAgent,
+          boulderAgentExplicitlySet: boulderState.agent !== undefined,
        })
        return
      }
@@ -120,6 +113,7 @@ export function createAtlasEventHandler(input: {
        return
      }

+      const now = Date.now()
      if (state.lastContinuationInjectedAt && now - state.lastContinuationInjectedAt < CONTINUATION_COOLDOWN_MS) {
        log(`[${HOOK_NAME}] Skipped: continuation cooldown active`, {
          sessionID,
@@ -138,7 +132,6 @@ export function createAtlasEventHandler(input: {
          remaining,
          total: progress.total,
          agent: boulderState.agent,
-          worktreePath: boulderState.worktree_path,
          backgroundManager,
          sessionState: state,
        })
--- a/src/hooks/atlas/index.test.ts
+++ b/src/hooks/atlas/index.test.ts
@@ -933,8 +933,8 @@ describe("atlas hook", () => {
      expect(callArgs.body.parts[0].text).toContain("2 remaining")
    })

-     test("should inject when last agent is sisyphus and boulder targets atlas explicitly", async () => {
-       // given - boulder explicitly set to atlas, but last agent is sisyphus (initial state after /start-work)
+     test("should not inject when last agent does not match boulder agent", async () => {
+       // given - boulder state with incomplete plan, but last agent does NOT match
       const planPath = join(TEST_DIR, "test-plan.md")
       writeFileSync(planPath, "# Plan\n- [ ] Task 1\n- [ ] Task 2")

@@ -947,7 +947,7 @@ describe("atlas hook", () => {
       }
       writeBoulderState(TEST_DIR, state)

-       // given - last agent is sisyphus (typical state right after /start-work)
+       // given - last agent is NOT the boulder agent
       cleanupMessageStorage(MAIN_SESSION_ID)
       setupMessageStorage(MAIN_SESSION_ID, "sisyphus")

@@ -962,39 +962,7 @@ describe("atlas hook", () => {
         },
       })

-       // then - should call prompt because sisyphus is always allowed for atlas boulders
-       expect(mockInput._promptMock).toHaveBeenCalled()
-     })
-
-     test("should not inject when last agent is non-sisyphus and does not match boulder agent", async () => {
-       // given - boulder explicitly set to atlas, last agent is hephaestus (unrelated agent)
-       const planPath = join(TEST_DIR, "test-plan.md")
-       writeFileSync(planPath, "# Plan\n- [ ] Task 1\n- [ ] Task 2")
-
-       const state: BoulderState = {
-         active_plan: planPath,
-         started_at: "2026-01-02T10:00:00Z",
-         session_ids: [MAIN_SESSION_ID],
-         plan_name: "test-plan",
-         agent: "atlas",
-       }
-       writeBoulderState(TEST_DIR, state)
-
-       cleanupMessageStorage(MAIN_SESSION_ID)
-       setupMessageStorage(MAIN_SESSION_ID, "hephaestus")
-
-       const mockInput = createMockPluginInput()
-       const hook = createAtlasHook(mockInput)
-
-       // when
-       await hook.handler({
-         event: {
-           type: "session.idle",
-           properties: { sessionID: MAIN_SESSION_ID },
-         },
-       })
-
-       // then - should NOT call prompt because hephaestus does not match atlas or sisyphus
+       // then - should NOT call prompt because agent does not match
       expect(mockInput._promptMock).not.toHaveBeenCalled()
     })

@@ -1154,144 +1122,6 @@ describe("atlas hook", () => {
      }
    })

-    test("should keep skipping continuation during 5-minute backoff after 2 consecutive failures", async () => {
-      //#given - boulder state with incomplete plan and prompt always fails
-      const planPath = join(TEST_DIR, "test-plan.md")
-      writeFileSync(planPath, "# Plan\n- [ ] Task 1\n- [ ] Task 2")
-
-      const state: BoulderState = {
-        active_plan: planPath,
-        started_at: "2026-01-02T10:00:00Z",
-        session_ids: [MAIN_SESSION_ID],
-        plan_name: "test-plan",
-      }
-      writeBoulderState(TEST_DIR, state)
-
-      const promptMock = mock(() => Promise.reject(new Error("Bad Request")))
-      const mockInput = createMockPluginInput({ promptMock })
-      const hook = createAtlasHook(mockInput)
-
-      const originalDateNow = Date.now
-      let now = 0
-      Date.now = () => now
-
-      try {
-        //#when - third idle occurs inside 5-minute backoff window
-        await hook.handler({ event: { type: "session.idle", properties: { sessionID: MAIN_SESSION_ID } } })
-        await flushMicrotasks()
-        now += 6000
-
-        await hook.handler({ event: { type: "session.idle", properties: { sessionID: MAIN_SESSION_ID } } })
-        await flushMicrotasks()
-        now += 60000
-
-        await hook.handler({ event: { type: "session.idle", properties: { sessionID: MAIN_SESSION_ID } } })
-        await flushMicrotasks()
-
-        //#then - third attempt should still be skipped
-        expect(promptMock).toHaveBeenCalledTimes(2)
-      } finally {
-        Date.now = originalDateNow
-      }
-    })
-
-    test("should retry continuation after 5-minute backoff expires following 2 consecutive failures", async () => {
-      //#given - boulder state with incomplete plan and prompt always fails
-      const planPath = join(TEST_DIR, "test-plan.md")
-      writeFileSync(planPath, "# Plan\n- [ ] Task 1\n- [ ] Task 2")
-
-      const state: BoulderState = {
-        active_plan: planPath,
-        started_at: "2026-01-02T10:00:00Z",
-        session_ids: [MAIN_SESSION_ID],
-        plan_name: "test-plan",
-      }
-      writeBoulderState(TEST_DIR, state)
-
-      const promptMock = mock(() => Promise.reject(new Error("Bad Request")))
-      const mockInput = createMockPluginInput({ promptMock })
-      const hook = createAtlasHook(mockInput)
-
-      const originalDateNow = Date.now
-      let now = 0
-      Date.now = () => now
-
-      try {
-        //#when - third idle occurs after 5+ minutes
-        await hook.handler({ event: { type: "session.idle", properties: { sessionID: MAIN_SESSION_ID } } })
-        await flushMicrotasks()
-        now += 6000
-
-        await hook.handler({ event: { type: "session.idle", properties: { sessionID: MAIN_SESSION_ID } } })
-        await flushMicrotasks()
-        now += 300000
-
-        await hook.handler({ event: { type: "session.idle", properties: { sessionID: MAIN_SESSION_ID } } })
-        await flushMicrotasks()
-
-        //#then - third attempt should run after backoff expiration
-        expect(promptMock).toHaveBeenCalledTimes(3)
-      } finally {
-        Date.now = originalDateNow
-      }
-    })
-
-    test("should reset prompt failure counter after successful retry beyond backoff window", async () => {
-      //#given - boulder state with incomplete plan and success on first retry after backoff
-      const planPath = join(TEST_DIR, "test-plan.md")
-      writeFileSync(planPath, "# Plan\n- [ ] Task 1\n- [ ] Task 2")
-
-      const state: BoulderState = {
-        active_plan: planPath,
-        started_at: "2026-01-02T10:00:00Z",
-        session_ids: [MAIN_SESSION_ID],
-        plan_name: "test-plan",
-      }
-      writeBoulderState(TEST_DIR, state)
-
-      const promptMock = mock((): Promise<void> => Promise.reject(new Error("Bad Request")))
-      promptMock.mockImplementationOnce(() => Promise.reject(new Error("Bad Request")))
-      promptMock.mockImplementationOnce(() => Promise.reject(new Error("Bad Request")))
-      promptMock.mockImplementationOnce(() => Promise.resolve(undefined))
-      const mockInput = createMockPluginInput({ promptMock })
-      const hook = createAtlasHook(mockInput)
-
-      const originalDateNow = Date.now
-      let now = 0
-      Date.now = () => now
-
-      try {
-        //#when - fail twice, recover after backoff with success, then fail twice again
-        await hook.handler({ event: { type: "session.idle", properties: { sessionID: MAIN_SESSION_ID } } })
-        await flushMicrotasks()
-        now += 6000
-
-        await hook.handler({ event: { type: "session.idle", properties: { sessionID: MAIN_SESSION_ID } } })
-        await flushMicrotasks()
-        now += 300000
-
-        await hook.handler({ event: { type: "session.idle", properties: { sessionID: MAIN_SESSION_ID } } })
-        await flushMicrotasks()
-        now += 6000
-
-        await hook.handler({ event: { type: "session.idle", properties: { sessionID: MAIN_SESSION_ID } } })
-        await flushMicrotasks()
-        now += 6000
-
-        await hook.handler({ event: { type: "session.idle", properties: { sessionID: MAIN_SESSION_ID } } })
-        await flushMicrotasks()
-        now += 6000
-
-        await hook.handler({ event: { type: "session.idle", properties: { sessionID: MAIN_SESSION_ID } } })
-        await flushMicrotasks()
-
-        //#then - success retry resets counter, so two additional failures are allowed before skip
-        expect(promptMock).toHaveBeenCalledTimes(5)
-      } finally {
-        Date.now = originalDateNow
-      }
-    })
-
    test("should reset continuation failure state on session.compacted event", async () => {
      //#given - boulder state with incomplete plan and prompt always fails
      const planPath = join(TEST_DIR, "test-plan.md")
--- a/src/hooks/atlas/system-reminder-templates.ts
+++ b/src/hooks/atlas/system-reminder-templates.ts
@@ -104,65 +104,6 @@ ALL three must be YES. "Probably" = NO. "I think so" = NO. Investigate until CER

 **DO NOT proceed to the next task until all 4 phases are complete and the gate passes.**`

-export const VERIFICATION_REMINDER_GEMINI = `**THE SUBAGENT HAS FINISHED. THEIR WORK IS EXTREMELY SUSPICIOUS.**
-
-The subagent CLAIMS this task is done. Based on thousands of executions, subagent claims are FALSE more often than true.
-They ROUTINELY:
- Ship code with syntax errors they didn't bother to check
- Create stub implementations with TODOs and call it "done"
- Write tests that pass trivially (testing nothing meaningful)
- Implement logic that does NOT match what was requested
- Add features nobody asked for and call it "improvement"
- Report "all tests pass" when they didn't run any tests
-
-**This is NOT a theoretical warning. This WILL happen on this task. Assume the work is BROKEN.**
-
-**YOU MUST VERIFY WITH ACTUAL TOOL CALLS. NOT REASONING. TOOL CALLS.**
-Thinking "it looks correct" is NOT verification. Running \`lsp_diagnostics\` IS.
-
---
-
-**PHASE 1: READ THE CODE FIRST (DO NOT SKIP — DO NOT RUN TESTS YET)**
-
-Read the code FIRST so you know what you're testing.
-
-1. \`Bash("git diff --stat")\` — see exactly which files changed.
-2. \`Read\` EVERY changed file — no exceptions, no skimming.
-3. For EACH file:
-   - Does this code ACTUALLY do what the task required? RE-READ the task spec.
-   - Any stubs, TODOs, placeholders? \`Grep\` for TODO, FIXME, HACK, xxx
-   - Anti-patterns? \`Grep\` for \`as any\`, \`@ts-ignore\`, empty catch
-   - Scope creep? Did the subagent add things NOT in the task spec?
-4. Cross-check EVERY claim against actual code.
-
-**If you cannot explain what every changed line does, GO BACK AND READ AGAIN.**
-
-**PHASE 2: RUN AUTOMATED CHECKS**
-
-1. \`lsp_diagnostics\` on EACH changed file — ZERO new errors. ACTUALLY RUN THIS.
-2. Run tests for changed modules, then full suite. ACTUALLY RUN THESE.
-3. Build/typecheck — exit 0.
-
-If Phase 1 found issues but Phase 2 passes: Phase 2 is WRONG. Fix the code.
-
-**PHASE 3: HANDS-ON QA (MANDATORY for user-facing changes)**
-
- **Frontend/UI**: \`/playwright\`
- **TUI/CLI**: \`interactive_bash\`
- **API/Backend**: \`Bash\` with curl
-
-**If user-facing and you did not run it, you are shipping UNTESTED BROKEN work.**
-
-**PHASE 4: GATE DECISION**
-
-1. Can I explain what EVERY changed line does? (If no → Phase 1)
-2. Did I SEE it work via tool calls? (If user-facing and no → Phase 3)
-3. Am I confident nothing is broken? (If no → broader tests)
-
-ALL three must be YES. "Probably" = NO. "I think so" = NO.
-
-**DO NOT proceed to the next task until all 4 phases are complete.**`
-
 export const ORCHESTRATOR_DELEGATION_REQUIRED = `

 ---
--- a/Show More
+++ b/Show More
				`@@ -0,0 +1 @@`
				`export { getMessageDir } from "../../shared"`