diff --git a/web/scripts/browser-flows.sh b/web/scripts/browser-flows.sh index 0c9abb6..c8b0397 100644 --- a/web/scripts/browser-flows.sh +++ b/web/scripts/browser-flows.sh @@ -3,12 +3,32 @@ set -euo pipefail BASE_URL="${BASE_URL:-${1:-http://127.0.0.1:3002}}" LOCALE="${LOCALE:-zh}" +SESSION_NAME="${SESSION_NAME:-learn-claude-code-flows-${LOCALE}}" ROOT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)" source "$ROOT_DIR/scripts/browser-test-lib.sh" +agent-browser() { + command agent-browser --session-name "$SESSION_NAME" "$@" +} + trap 'stop_static_server_if_started; agent-browser close >/dev/null 2>&1 || true' EXIT +locale_text() { + local key="$1" + case "$LOCALE:$key" in + zh:deep_dive) echo '深入探索' ;; + en:deep_dive) echo 'Deep Dive' ;; + ja:deep_dive) echo '深掘り' ;; + + zh:bridge_control_plane) echo '工具控制平面' ;; + en:bridge_control_plane) echo 'Tool Control Plane' ;; + ja:bridge_control_plane) echo 'ツール制御プレーン' ;; + + *) echo "Unknown locale text key: ${LOCALE}:${key}" >&2; return 1 ;; + esac +} + wait_page() { agent-browser wait --load networkidle >/dev/null 2>&1 || agent-browser wait 600 >/dev/null 2>&1 || true agent-browser wait 1200 >/dev/null 2>&1 || true @@ -171,7 +191,7 @@ flow_home_to_s01() { click_link_by_href "/${LOCALE}/s01/" wait_page assert_url_contains "/${LOCALE}/s01/" - assert_body_contains 'Agent 循环' + assert_body_contains 's01' assert_no_overflow assert_no_page_errors } @@ -179,7 +199,8 @@ flow_home_to_s01() { flow_home_to_timeline() { open_page "/${LOCALE}/timeline/" assert_url_contains "/${LOCALE}/timeline/" - assert_body_contains '按 4 个阶段渐进搭建' + assert_body_contains 's01' + assert_body_contains 's19' assert_no_overflow assert_no_page_errors } @@ -187,68 +208,69 @@ flow_home_to_timeline() { flow_home_to_layers() { open_page "/${LOCALE}/layers/" assert_url_contains "/${LOCALE}/layers/" - assert_body_contains '阶段入口' + assert_body_contains 'P1' + assert_body_contains 's19' assert_no_overflow assert_no_page_errors } flow_home_to_compare() { open_page "/${LOCALE}/" - click_link_by_href "/${LOCALE}/compare/" '版本对比' + click_link_by_href "/${LOCALE}/compare/" wait_page assert_url_contains "/${LOCALE}/compare/" - assert_body_contains '学习路径对比' + assert_body_contains 's14 -> s15' assert_no_overflow assert_no_page_errors } flow_compare_default_state() { open_page "/${LOCALE}/compare" - assert_body_contains '跃迁诊断' - assert_body_contains 'Agent 循环' - assert_body_contains '工具使用' + assert_body_contains 's01' + assert_body_contains 's02' + assert_body_contains 's14 -> s15' assert_no_overflow assert_no_page_errors } flow_timeline_to_stage_exit() { open_page "/${LOCALE}/timeline" - click_link_exact '打开阶段收口: s06' + click_link_by_href "/${LOCALE}/s06/" wait_page assert_url_contains "/${LOCALE}/s06/" - assert_body_contains '上下文压缩' + assert_body_contains 's06' assert_no_overflow assert_no_page_errors } flow_layers_to_stage_entry() { open_page "/${LOCALE}/layers" - click_link_by_href "/${LOCALE}/s15/" '阶段入口' + click_link_by_href "/${LOCALE}/s15/" wait_page assert_url_contains "/${LOCALE}/s15/" - assert_body_contains 'Agent 团队' + assert_body_contains 's15' assert_no_overflow assert_no_page_errors } flow_chapter_to_bridge_doc() { open_page "/${LOCALE}/s02" - agent-browser --json find text '深入探索' click >/dev/null + agent-browser --json find text "$(locale_text deep_dive)" click >/dev/null wait_page - click_link_by_href "/${LOCALE}/docs/s02a-tool-control-plane/" '工具控制平面' + click_link_by_href "/${LOCALE}/docs/s02a-tool-control-plane/" "$(locale_text bridge_control_plane)" wait_page assert_url_contains "/${LOCALE}/docs/s02a-tool-control-plane/" - assert_body_contains '工具控制平面' + assert_body_contains "$(locale_text bridge_control_plane)" assert_no_overflow assert_no_page_errors } flow_bridge_doc_home_return() { open_page "/${LOCALE}/docs/s00f-code-reading-order" - click_link_by_href "/${LOCALE}/" '回到学习主线' + click_link_by_href "/${LOCALE}/" wait_page assert_url_contains "/${LOCALE}/" - assert_body_contains '开始学习' + assert_body_contains 's01' assert_no_overflow assert_no_page_errors } @@ -258,7 +280,7 @@ flow_bridge_doc_back_to_chapter() { click_link_by_href "/${LOCALE}/s02/" 's02' wait_page assert_url_contains "/${LOCALE}/s02/" - assert_body_contains '工具使用' + assert_body_contains 's02' assert_no_overflow assert_no_page_errors } @@ -281,19 +303,18 @@ flow_compare_preset() { open_page "/${LOCALE}/compare" agent-browser --json find text 's14 -> s15' click >/dev/null agent-browser wait 800 >/dev/null 2>&1 || true - assert_body_contains '跃迁诊断' - assert_body_contains 'Agent 团队' - assert_body_contains '更稳的读法' + assert_body_contains 's14' + assert_body_contains 's15' assert_no_overflow assert_no_page_errors } flow_chapter_next_navigation() { open_page "/${LOCALE}/s15" - click_link_by_href "/${LOCALE}/s16/" '下一章' + click_link_by_href "/${LOCALE}/s16/" wait_page assert_url_contains "/${LOCALE}/s16/" - assert_body_contains '团队协议' + assert_body_contains 's16' assert_no_overflow assert_no_page_errors } diff --git a/web/src/data/generated/docs.json b/web/src/data/generated/docs.json index 57927d1..17ec652 100644 --- a/web/src/data/generated/docs.json +++ b/web/src/data/generated/docs.json @@ -870,7 +870,7 @@ "title": "s12: Task System", "kind": "chapter", "filename": "s12-task-system.md", - "content": "# s12: Task System\n\n`s01 > s02 > s03 > s04 > s05 > s06 | s07 > s08 > s09 > s10 > s11 > [ s12 ]`\n\n> *\"大きな目標を小タスクに分解し、順序付けし、ディスクに記録する\"* -- ファイルベースのタスクグラフ、マルチエージェント協調の基盤。\n>\n> **Harness 層**: 永続タスク -- どの会話よりも長く生きる目標。\n\n## 問題\n\ns03のTodoManagerはメモリ上のフラットなチェックリストに過ぎない: 順序なし、依存関係なし、ステータスは完了か未完了のみ。実際の目標には構造がある -- タスクBはタスクAに依存し、タスクCとDは並行実行でき、タスクEはCとDの両方を待つ。\n\n明示的な関係がなければ、エージェントは何が実行可能で、何がブロックされ、何が同時に走れるかを判断できない。しかもリストはメモリ上にしかないため、コンテキスト圧縮(s06)で消える。\n\n## 主線とどう併読するか\n\n- `s03` からそのまま来たなら、[`data-structures.md`](./data-structures.md) へ戻って `TodoItem` / `PlanState` と `TaskRecord` を分けます。\n- object 境界が混ざり始めたら、[`entity-map.md`](./entity-map.md) で message、task、runtime task、teammate を分離してから戻ります。\n- 次に `s13` を読むなら、[`s13a-runtime-task-model.md`](./s13a-runtime-task-model.md) を横に置いて、durable task と runtime task を同じ言葉で潰さないようにします。\n\n## 解決策\n\nフラットなチェックリストをディスクに永続化する**タスクグラフ**に昇格させる。各タスクは1つのJSONファイルで、ステータス・前方依存(`blockedBy`)・後方依存(`blocks`)を持つ。タスクグラフは常に3つの問いに答える:\n\n- **何が実行可能か?** -- `pending`ステータスで`blockedBy`が空のタスク。\n- **何がブロックされているか?** -- 未完了の依存を待つタスク。\n- **何が完了したか?** -- `completed`のタスク。完了時に後続タスクを自動的にアンブロックする。\n\n```\n.tasks/\n task_1.json {\"id\":1, \"status\":\"completed\"}\n task_2.json {\"id\":2, \"blockedBy\":[1], \"status\":\"pending\"}\n task_3.json {\"id\":3, \"blockedBy\":[1], \"status\":\"pending\"}\n task_4.json {\"id\":4, \"blockedBy\":[2,3], \"status\":\"pending\"}\n\nタスクグラフ (DAG):\n +----------+\n +--> | task 2 | --+\n | | pending | |\n+----------+ +----------+ +--> +----------+\n| task 1 | | task 4 |\n| completed| --> +----------+ +--> | blocked |\n+----------+ | task 3 | --+ +----------+\n | pending |\n +----------+\n\n順序: task 1 は 2 と 3 より先に完了する必要がある\n並行: task 2 と 3 は同時に実行できる\n依存: task 4 は 2 と 3 の両方を待つ\nステータス: pending -> in_progress -> completed\n```\n\nこのタスクグラフは後続の runtime / platform 章の協調バックボーンになる: バックグラウンド実行(`s13`)、マルチエージェントチーム(`s15+`)、worktree 分離(`s18`)はすべてこの durable な構造の恩恵を受ける。\n\n## 仕組み\n\n1. **TaskManager**: タスクごとに1つのJSONファイル、依存グラフ付きCRUD。\n\n```python\nclass TaskManager:\n def __init__(self, tasks_dir: Path):\n self.dir = tasks_dir\n self.dir.mkdir(exist_ok=True)\n self._next_id = self._max_id() + 1\n\n def create(self, subject, description=\"\"):\n task = {\"id\": self._next_id, \"subject\": subject,\n \"status\": \"pending\", \"blockedBy\": [],\n \"blocks\": [], \"owner\": \"\"}\n self._save(task)\n self._next_id += 1\n return json.dumps(task, indent=2)\n```\n\n2. **依存解除**: タスク完了時に、他タスクの`blockedBy`リストから完了IDを除去し、後続タスクをアンブロックする。\n\n```python\ndef _clear_dependency(self, completed_id):\n for f in self.dir.glob(\"task_*.json\"):\n task = json.loads(f.read_text())\n if completed_id in task.get(\"blockedBy\", []):\n task[\"blockedBy\"].remove(completed_id)\n self._save(task)\n```\n\n3. **ステータス遷移 + 依存配線**: `update`がステータス変更と依存エッジを担う。\n\n```python\ndef update(self, task_id, status=None,\n add_blocked_by=None, add_blocks=None):\n task = self._load(task_id)\n if status:\n task[\"status\"] = status\n if status == \"completed\":\n self._clear_dependency(task_id)\n self._save(task)\n```\n\n4. 4つのタスクツールをディスパッチマップに追加する。\n\n```python\nTOOL_HANDLERS = {\n # ...base tools...\n \"task_create\": lambda **kw: TASKS.create(kw[\"subject\"]),\n \"task_update\": lambda **kw: TASKS.update(kw[\"task_id\"], kw.get(\"status\")),\n \"task_list\": lambda **kw: TASKS.list_all(),\n \"task_get\": lambda **kw: TASKS.get(kw[\"task_id\"]),\n}\n```\n\n`s12` 以降、タスクグラフが durable なマルチステップ作業のデフォルトになる。`s03` の Todo は軽量な単一セッション用チェックリストとして残る。\n\n## s06からの変更点\n\n| コンポーネント | Before (s06) | After (s12) |\n|---|---|---|\n| Tools | 5 | 8 (`task_create/update/list/get`) |\n| 計画モデル | フラットチェックリスト (メモリ) | 依存関係付きタスクグラフ (ディスク) |\n| 関係 | なし | `blockedBy` + `blocks` エッジ |\n| ステータス追跡 | 完了か未完了 | `pending` -> `in_progress` -> `completed` |\n| 永続性 | 圧縮で消失 | 圧縮・再起動後も存続 |\n\n## 試してみる\n\n```sh\ncd learn-claude-code\npython agents/s12_task_system.py\n```\n\n1. `Create 3 tasks: \"Setup project\", \"Write code\", \"Write tests\". Make them depend on each other in order.`\n2. `List all tasks and show the dependency graph`\n3. `Complete task 1 and then list tasks to see task 2 unblocked`\n4. `Create a task board for refactoring: parse -> transform -> emit -> test, where transform and emit can run in parallel after parse`\n\n## 教学上の境界\n\nこのリポジトリで本当に重要なのは、完全な製品向け保存層の再現ではありません。\n\n重要なのは:\n\n- durable なタスク記録\n- 明示的な依存エッジ\n- 分かりやすい状態遷移\n- 後続章が再利用できる構造\n\nこの 4 点を自分で実装できれば、タスクシステムの核心はつかめています。\n" + "content": "# s12: Task System\n\n`s01 > s02 > s03 > s04 > s05 > s06 | s07 > s08 > s09 > s10 > s11 > [ s12 ]`\n\n> *\"大きな目標を小タスクに分解し、順序付けし、ディスクに記録する\"* -- ファイルベースのタスクグラフ、マルチエージェント協調の基盤。\n>\n> **Harness 層**: 永続タスク -- どの会話よりも長く生きる目標。\n\n## 問題\n\ns03のTodoManagerはメモリ上のフラットなチェックリストに過ぎない: 順序なし、依存関係なし、ステータスは完了か未完了のみ。実際の目標には構造がある -- タスクBはタスクAに依存し、タスクCとDは並行実行でき、タスクEはCとDの両方を待つ。\n\n明示的な関係がなければ、エージェントは何が実行可能で、何がブロックされ、何が同時に走れるかを判断できない。しかもリストはメモリ上にしかないため、コンテキスト圧縮(s06)で消える。\n\n## 主線とどう併読するか\n\n- `s03` からそのまま来たなら、[`data-structures.md`](./data-structures.md) へ戻って `TodoItem` / `PlanState` と `TaskRecord` を分けます。\n- object 境界が混ざり始めたら、[`entity-map.md`](./entity-map.md) で message、task、runtime task、teammate を分離してから戻ります。\n- 次に `s13` を読むなら、[`s13a-runtime-task-model.md`](./s13a-runtime-task-model.md) を横に置いて、durable task と runtime task を同じ言葉で潰さないようにします。\n\n## 解決策\n\nフラットなチェックリストをディスクに永続化する**タスクグラフ**に昇格させる。各タスクは1つのJSONファイルで、ステータス・前方依存(`blockedBy`)を持つ。タスクグラフは常に3つの問いに答える:\n\n- **何が実行可能か?** -- `pending`ステータスで`blockedBy`が空のタスク。\n- **何がブロックされているか?** -- 未完了の依存を待つタスク。\n- **何が完了したか?** -- `completed`のタスク。完了時に後続タスクを自動的にアンブロックする。\n\n```\n.tasks/\n task_1.json {\"id\":1, \"status\":\"completed\"}\n task_2.json {\"id\":2, \"blockedBy\":[1], \"status\":\"pending\"}\n task_3.json {\"id\":3, \"blockedBy\":[1], \"status\":\"pending\"}\n task_4.json {\"id\":4, \"blockedBy\":[2,3], \"status\":\"pending\"}\n\nタスクグラフ (DAG):\n +----------+\n +--> | task 2 | --+\n | | pending | |\n+----------+ +----------+ +--> +----------+\n| task 1 | | task 4 |\n| completed| --> +----------+ +--> | blocked |\n+----------+ | task 3 | --+ +----------+\n | pending |\n +----------+\n\n順序: task 1 は 2 と 3 より先に完了する必要がある\n並行: task 2 と 3 は同時に実行できる\n依存: task 4 は 2 と 3 の両方を待つ\nステータス: pending -> in_progress -> completed\n```\n\nこのタスクグラフは後続の runtime / platform 章の協調バックボーンになる: バックグラウンド実行(`s13`)、マルチエージェントチーム(`s15+`)、worktree 分離(`s18`)はすべてこの durable な構造の恩恵を受ける。\n\n## 仕組み\n\n1. **TaskManager**: タスクごとに1つのJSONファイル、依存グラフ付きCRUD。\n\n```python\nclass TaskManager:\n def __init__(self, tasks_dir: Path):\n self.dir = tasks_dir\n self.dir.mkdir(exist_ok=True)\n self._next_id = self._max_id() + 1\n\n def create(self, subject, description=\"\"):\n task = {\"id\": self._next_id, \"subject\": subject,\n \"status\": \"pending\", \"blockedBy\": [],\n \"owner\": \"\"}\n self._save(task)\n self._next_id += 1\n return json.dumps(task, indent=2)\n```\n\n2. **依存解除**: タスク完了時に、他タスクの`blockedBy`リストから完了IDを除去し、後続タスクをアンブロックする。\n\n```python\ndef _clear_dependency(self, completed_id):\n for f in self.dir.glob(\"task_*.json\"):\n task = json.loads(f.read_text())\n if completed_id in task.get(\"blockedBy\", []):\n task[\"blockedBy\"].remove(completed_id)\n self._save(task)\n```\n\n3. **ステータス遷移 + 依存配線**: `update`がステータス変更と依存エッジを担う。\n\n```python\ndef update(self, task_id, status=None,\n add_blocked_by=None, remove_blocked_by=None):\n task = self._load(task_id)\n if status:\n task[\"status\"] = status\n if status == \"completed\":\n self._clear_dependency(task_id)\n if add_blocked_by:\n task[\"blockedBy\"] = list(set(task[\"blockedBy\"] + add_blocked_by))\n if remove_blocked_by:\n task[\"blockedBy\"] = [x for x in task[\"blockedBy\"] if x not in remove_blocked_by]\n self._save(task)\n```\n\n4. 4つのタスクツールをディスパッチマップに追加する。\n\n```python\nTOOL_HANDLERS = {\n # ...base tools...\n \"task_create\": lambda **kw: TASKS.create(kw[\"subject\"]),\n \"task_update\": lambda **kw: TASKS.update(kw[\"task_id\"], kw.get(\"status\")),\n \"task_list\": lambda **kw: TASKS.list_all(),\n \"task_get\": lambda **kw: TASKS.get(kw[\"task_id\"]),\n}\n```\n\n`s12` 以降、タスクグラフが durable なマルチステップ作業のデフォルトになる。`s03` の Todo は軽量な単一セッション用チェックリストとして残る。\n\n## s06からの変更点\n\n| コンポーネント | Before (s06) | After (s12) |\n|---|---|---|\n| Tools | 5 | 8 (`task_create/update/list/get`) |\n| 計画モデル | フラットチェックリスト (メモリ) | 依存関係付きタスクグラフ (ディスク) |\n| 関係 | なし | `blockedBy` エッジ |\n| ステータス追跡 | 完了か未完了 | `pending` -> `in_progress` -> `completed` |\n| 永続性 | 圧縮で消失 | 圧縮・再起動後も存続 |\n\n## 試してみる\n\n```sh\ncd learn-claude-code\npython agents/s12_task_system.py\n```\n\n1. `Create 3 tasks: \"Setup project\", \"Write code\", \"Write tests\". Make them depend on each other in order.`\n2. `List all tasks and show the dependency graph`\n3. `Complete task 1 and then list tasks to see task 2 unblocked`\n4. `Create a task board for refactoring: parse -> transform -> emit -> test, where transform and emit can run in parallel after parse`\n\n## 教学上の境界\n\nこのリポジトリで本当に重要なのは、完全な製品向け保存層の再現ではありません。\n\n重要なのは:\n\n- durable なタスク記録\n- 明示的な依存エッジ\n- 分かりやすい状態遷移\n- 後続章が再利用できる構造\n\nこの 4 点を自分で実装できれば、タスクシステムの核心はつかめています。\n" }, { "version": "s13", diff --git a/web/src/data/generated/versions.json b/web/src/data/generated/versions.json index 0399a60..a33fca6 100644 --- a/web/src/data/generated/versions.json +++ b/web/src/data/generated/versions.json @@ -5,7 +5,7 @@ "filename": "s01_agent_loop.py", "title": "The Agent Loop", "subtitle": "Minimal Closed Loop", - "loc": 119, + "loc": 130, "tools": [ "bash" ], @@ -17,46 +17,46 @@ "classes": [ { "name": "LoopState", - "startLine": 50, - "endLine": 56 + "startLine": 61, + "endLine": 67 } ], "functions": [ { "name": "run_bash", "signature": "def run_bash(command: str)", - "startLine": 57 + "startLine": 68 }, { "name": "extract_text", "signature": "def extract_text(content)", - "startLine": 77 + "startLine": 90 }, { "name": "execute_tool_calls", "signature": "def execute_tool_calls(response_content)", - "startLine": 88 + "startLine": 101 }, { "name": "run_one_turn", "signature": "def run_one_turn(state: LoopState)", - "startLine": 105 + "startLine": 118 }, { "name": "agent_loop", "signature": "def agent_loop(state: LoopState)", - "startLine": 130 + "startLine": 143 } ], "layer": "core", - "source": "#!/usr/bin/env python3\n# Harness: the loop -- keep feeding real tool results back into the model.\n\"\"\"\ns01_agent_loop.py - The Agent Loop\n\nThis file teaches the smallest useful coding-agent pattern:\n\n user message\n -> model reply\n -> if tool_use: execute tools\n -> write tool_result back to messages\n -> continue\n\nIt intentionally keeps the loop small, but still makes the loop state explicit\nso later chapters can grow from the same structure.\n\"\"\"\n\nimport os\nimport subprocess\nfrom dataclasses import dataclass\n\nfrom anthropic import Anthropic\nfrom dotenv import load_dotenv\n\nload_dotenv(override=True)\n\nif os.getenv(\"ANTHROPIC_BASE_URL\"):\n os.environ.pop(\"ANTHROPIC_AUTH_TOKEN\", None)\n\nclient = Anthropic(base_url=os.getenv(\"ANTHROPIC_BASE_URL\"))\nMODEL = os.environ[\"MODEL_ID\"]\n\nSYSTEM = (\n f\"You are a coding agent at {os.getcwd()}. \"\n \"Use bash to inspect and change the workspace. Act first, then report clearly.\"\n)\n\nTOOLS = [{\n \"name\": \"bash\",\n \"description\": \"Run a shell command in the current workspace.\",\n \"input_schema\": {\n \"type\": \"object\",\n \"properties\": {\"command\": {\"type\": \"string\"}},\n \"required\": [\"command\"],\n },\n}]\n\n\n@dataclass\nclass LoopState:\n # The minimal loop state: history, loop count, and why we continue.\n messages: list\n turn_count: int = 1\n transition_reason: str | None = None\n\n\ndef run_bash(command: str) -> str:\n dangerous = [\"rm -rf /\", \"sudo\", \"shutdown\", \"reboot\", \"> /dev/\"]\n if any(item in command for item in dangerous):\n return \"Error: Dangerous command blocked\"\n try:\n result = subprocess.run(\n command,\n shell=True,\n cwd=os.getcwd(),\n capture_output=True,\n text=True,\n timeout=120,\n )\n except subprocess.TimeoutExpired:\n return \"Error: Timeout (120s)\"\n\n output = (result.stdout + result.stderr).strip()\n return output[:50000] if output else \"(no output)\"\n\n\ndef extract_text(content) -> str:\n if not isinstance(content, list):\n return \"\"\n texts = []\n for block in content:\n text = getattr(block, \"text\", None)\n if text:\n texts.append(text)\n return \"\\n\".join(texts).strip()\n\n\ndef execute_tool_calls(response_content) -> list[dict]:\n results = []\n for block in response_content:\n if block.type != \"tool_use\":\n continue\n command = block.input[\"command\"]\n print(f\"\\033[33m$ {command}\\033[0m\")\n output = run_bash(command)\n print(output[:200])\n results.append({\n \"type\": \"tool_result\",\n \"tool_use_id\": block.id,\n \"content\": output,\n })\n return results\n\n\ndef run_one_turn(state: LoopState) -> bool:\n response = client.messages.create(\n model=MODEL,\n system=SYSTEM,\n messages=state.messages,\n tools=TOOLS,\n max_tokens=8000,\n )\n state.messages.append({\"role\": \"assistant\", \"content\": response.content})\n\n if response.stop_reason != \"tool_use\":\n state.transition_reason = None\n return False\n\n results = execute_tool_calls(response.content)\n if not results:\n state.transition_reason = None\n return False\n\n state.messages.append({\"role\": \"user\", \"content\": results})\n state.turn_count += 1\n state.transition_reason = \"tool_result\"\n return True\n\n\ndef agent_loop(state: LoopState) -> None:\n while run_one_turn(state):\n pass\n\n\nif __name__ == \"__main__\":\n history = []\n while True:\n try:\n query = input(\"\\033[36ms01 >> \\033[0m\")\n except (EOFError, KeyboardInterrupt):\n break\n if query.strip().lower() in (\"q\", \"exit\", \"\"):\n break\n\n history.append({\"role\": \"user\", \"content\": query})\n state = LoopState(messages=history)\n agent_loop(state)\n\n final_text = extract_text(history[-1][\"content\"])\n if final_text:\n print(final_text)\n print()\n" + "source": "#!/usr/bin/env python3\n# Harness: the loop -- keep feeding real tool results back into the model.\n\"\"\"\ns01_agent_loop.py - The Agent Loop\n\nThis file teaches the smallest useful coding-agent pattern:\n\n user message\n -> model reply\n -> if tool_use: execute tools\n -> write tool_result back to messages\n -> continue\n\nIt intentionally keeps the loop small, but still makes the loop state explicit\nso later chapters can grow from the same structure.\n\"\"\"\n\nimport os\nimport subprocess\nfrom dataclasses import dataclass\n\ntry:\n import readline\n # #143 UTF-8 backspace fix for macOS libedit\n readline.parse_and_bind('set bind-tty-special-chars off')\n readline.parse_and_bind('set input-meta on')\n readline.parse_and_bind('set output-meta on')\n readline.parse_and_bind('set convert-meta off')\n readline.parse_and_bind('set enable-meta-keybindings on')\nexcept ImportError:\n pass\n\nfrom anthropic import Anthropic\nfrom dotenv import load_dotenv\n\nload_dotenv(override=True)\n\nif os.getenv(\"ANTHROPIC_BASE_URL\"):\n os.environ.pop(\"ANTHROPIC_AUTH_TOKEN\", None)\n\nclient = Anthropic(base_url=os.getenv(\"ANTHROPIC_BASE_URL\"))\nMODEL = os.environ[\"MODEL_ID\"]\n\nSYSTEM = (\n f\"You are a coding agent at {os.getcwd()}. \"\n \"Use bash to inspect and change the workspace. Act first, then report clearly.\"\n)\n\nTOOLS = [{\n \"name\": \"bash\",\n \"description\": \"Run a shell command in the current workspace.\",\n \"input_schema\": {\n \"type\": \"object\",\n \"properties\": {\"command\": {\"type\": \"string\"}},\n \"required\": [\"command\"],\n },\n}]\n\n\n@dataclass\nclass LoopState:\n # The minimal loop state: history, loop count, and why we continue.\n messages: list\n turn_count: int = 1\n transition_reason: str | None = None\n\n\ndef run_bash(command: str) -> str:\n dangerous = [\"rm -rf /\", \"sudo\", \"shutdown\", \"reboot\", \"> /dev/\"]\n if any(item in command for item in dangerous):\n return \"Error: Dangerous command blocked\"\n try:\n result = subprocess.run(\n command,\n shell=True,\n cwd=os.getcwd(),\n capture_output=True,\n text=True,\n timeout=120,\n )\n except subprocess.TimeoutExpired:\n return \"Error: Timeout (120s)\"\n except (FileNotFoundError, OSError) as e:\n return f\"Error: {e}\"\n\n output = (result.stdout + result.stderr).strip()\n return output[:50000] if output else \"(no output)\"\n\n\ndef extract_text(content) -> str:\n if not isinstance(content, list):\n return \"\"\n texts = []\n for block in content:\n text = getattr(block, \"text\", None)\n if text:\n texts.append(text)\n return \"\\n\".join(texts).strip()\n\n\ndef execute_tool_calls(response_content) -> list[dict]:\n results = []\n for block in response_content:\n if block.type != \"tool_use\":\n continue\n command = block.input[\"command\"]\n print(f\"\\033[33m$ {command}\\033[0m\")\n output = run_bash(command)\n print(output[:200])\n results.append({\n \"type\": \"tool_result\",\n \"tool_use_id\": block.id,\n \"content\": output,\n })\n return results\n\n\ndef run_one_turn(state: LoopState) -> bool:\n response = client.messages.create(\n model=MODEL,\n system=SYSTEM,\n messages=state.messages,\n tools=TOOLS,\n max_tokens=8000,\n )\n state.messages.append({\"role\": \"assistant\", \"content\": response.content})\n\n if response.stop_reason != \"tool_use\":\n state.transition_reason = None\n return False\n\n results = execute_tool_calls(response.content)\n if not results:\n state.transition_reason = None\n return False\n\n state.messages.append({\"role\": \"user\", \"content\": results})\n state.turn_count += 1\n state.transition_reason = \"tool_result\"\n return True\n\n\ndef agent_loop(state: LoopState) -> None:\n while run_one_turn(state):\n pass\n\n\nif __name__ == \"__main__\":\n history = []\n while True:\n try:\n query = input(\"\\033[36ms01 >> \\033[0m\")\n except (EOFError, KeyboardInterrupt):\n break\n if query.strip().lower() in (\"q\", \"exit\", \"\"):\n break\n\n history.append({\"role\": \"user\", \"content\": query})\n state = LoopState(messages=history)\n agent_loop(state)\n\n final_text = extract_text(history[-1][\"content\"])\n if final_text:\n print(final_text)\n print()\n" }, { "id": "s02", "filename": "s02_tool_use.py", "title": "Tool Use", "subtitle": "Route Intent into Action", - "loc": 168, + "loc": 169, "tools": [ "bash", "read_file", @@ -109,7 +109,7 @@ } ], "layer": "core", - "source": "#!/usr/bin/env python3\n# Harness: tool dispatch -- expanding what the model can reach.\n\"\"\"\ns02_tool_use.py - Tool dispatch + message normalization\n\nThe agent loop from s01 didn't change. We added tools to the dispatch map,\nand a normalize_messages() function that cleans up the message list before\neach API call.\n\nKey insight: \"The loop didn't change at all. I just added tools.\"\n\"\"\"\n\nimport os\nimport subprocess\nfrom pathlib import Path\n\nfrom anthropic import Anthropic\nfrom dotenv import load_dotenv\n\nload_dotenv(override=True)\n\nif os.getenv(\"ANTHROPIC_BASE_URL\"):\n os.environ.pop(\"ANTHROPIC_AUTH_TOKEN\", None)\n\nWORKDIR = Path.cwd()\nclient = Anthropic(base_url=os.getenv(\"ANTHROPIC_BASE_URL\"))\nMODEL = os.environ[\"MODEL_ID\"]\n\nSYSTEM = f\"You are a coding agent at {WORKDIR}. Use tools to solve tasks. Act, don't explain.\"\n\n\ndef safe_path(p: str) -> Path:\n path = (WORKDIR / p).resolve()\n if not path.is_relative_to(WORKDIR):\n raise ValueError(f\"Path escapes workspace: {p}\")\n return path\n\n\ndef run_bash(command: str) -> str:\n dangerous = [\"rm -rf /\", \"sudo\", \"shutdown\", \"reboot\", \"> /dev/\"]\n if any(d in command for d in dangerous):\n return \"Error: Dangerous command blocked\"\n try:\n r = subprocess.run(command, shell=True, cwd=WORKDIR,\n capture_output=True, text=True, timeout=120)\n out = (r.stdout + r.stderr).strip()\n return out[:50000] if out else \"(no output)\"\n except subprocess.TimeoutExpired:\n return \"Error: Timeout (120s)\"\n\n\ndef run_read(path: str, limit: int = None) -> str:\n try:\n text = safe_path(path).read_text()\n lines = text.splitlines()\n if limit and limit < len(lines):\n lines = lines[:limit] + [f\"... ({len(lines) - limit} more lines)\"]\n return \"\\n\".join(lines)[:50000]\n except Exception as e:\n return f\"Error: {e}\"\n\n\ndef run_write(path: str, content: str) -> str:\n try:\n fp = safe_path(path)\n fp.parent.mkdir(parents=True, exist_ok=True)\n fp.write_text(content)\n return f\"Wrote {len(content)} bytes to {path}\"\n except Exception as e:\n return f\"Error: {e}\"\n\n\ndef run_edit(path: str, old_text: str, new_text: str) -> str:\n try:\n fp = safe_path(path)\n content = fp.read_text()\n if old_text not in content:\n return f\"Error: Text not found in {path}\"\n fp.write_text(content.replace(old_text, new_text, 1))\n return f\"Edited {path}\"\n except Exception as e:\n return f\"Error: {e}\"\n\n\n# -- Concurrency safety classification --\n# Read-only tools can safely run in parallel; mutating tools must be serialized.\nCONCURRENCY_SAFE = {\"read_file\"}\nCONCURRENCY_UNSAFE = {\"write_file\", \"edit_file\"}\n\n# -- The dispatch map: {tool_name: handler} --\nTOOL_HANDLERS = {\n \"bash\": lambda **kw: run_bash(kw[\"command\"]),\n \"read_file\": lambda **kw: run_read(kw[\"path\"], kw.get(\"limit\")),\n \"write_file\": lambda **kw: run_write(kw[\"path\"], kw[\"content\"]),\n \"edit_file\": lambda **kw: run_edit(kw[\"path\"], kw[\"old_text\"], kw[\"new_text\"]),\n}\n\nTOOLS = [\n {\"name\": \"bash\", \"description\": \"Run a shell command.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"command\": {\"type\": \"string\"}}, \"required\": [\"command\"]}},\n {\"name\": \"read_file\", \"description\": \"Read file contents.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"limit\": {\"type\": \"integer\"}}, \"required\": [\"path\"]}},\n {\"name\": \"write_file\", \"description\": \"Write content to file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"content\": {\"type\": \"string\"}}, \"required\": [\"path\", \"content\"]}},\n {\"name\": \"edit_file\", \"description\": \"Replace exact text in file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"old_text\": {\"type\": \"string\"}, \"new_text\": {\"type\": \"string\"}}, \"required\": [\"path\", \"old_text\", \"new_text\"]}},\n]\n\n\ndef normalize_messages(messages: list) -> list:\n \"\"\"Clean up messages before sending to the API.\n\n Three jobs:\n 1. Strip internal metadata fields the API doesn't understand\n 2. Ensure every tool_use has a matching tool_result (insert placeholder if missing)\n 3. Merge consecutive same-role messages (API requires strict alternation)\n \"\"\"\n cleaned = []\n for msg in messages:\n clean = {\"role\": msg[\"role\"]}\n if isinstance(msg.get(\"content\"), str):\n clean[\"content\"] = msg[\"content\"]\n elif isinstance(msg.get(\"content\"), list):\n clean[\"content\"] = [\n {k: v for k, v in block.items()\n if not k.startswith(\"_\")}\n for block in msg[\"content\"]\n if isinstance(block, dict)\n ]\n else:\n clean[\"content\"] = msg.get(\"content\", \"\")\n cleaned.append(clean)\n\n # Collect existing tool_result IDs\n existing_results = set()\n for msg in cleaned:\n if isinstance(msg.get(\"content\"), list):\n for block in msg[\"content\"]:\n if isinstance(block, dict) and block.get(\"type\") == \"tool_result\":\n existing_results.add(block.get(\"tool_use_id\"))\n\n # Find orphaned tool_use blocks and insert placeholder results\n for msg in cleaned:\n if msg[\"role\"] != \"assistant\" or not isinstance(msg.get(\"content\"), list):\n continue\n for block in msg[\"content\"]:\n if not isinstance(block, dict):\n continue\n if block.get(\"type\") == \"tool_use\" and block.get(\"id\") not in existing_results:\n cleaned.append({\"role\": \"user\", \"content\": [\n {\"type\": \"tool_result\", \"tool_use_id\": block[\"id\"],\n \"content\": \"(cancelled)\"}\n ]})\n\n # Merge consecutive same-role messages\n if not cleaned:\n return cleaned\n merged = [cleaned[0]]\n for msg in cleaned[1:]:\n if msg[\"role\"] == merged[-1][\"role\"]:\n prev = merged[-1]\n prev_c = prev[\"content\"] if isinstance(prev[\"content\"], list) \\\n else [{\"type\": \"text\", \"text\": str(prev[\"content\"])}]\n curr_c = msg[\"content\"] if isinstance(msg[\"content\"], list) \\\n else [{\"type\": \"text\", \"text\": str(msg[\"content\"])}]\n prev[\"content\"] = prev_c + curr_c\n else:\n merged.append(msg)\n return merged\n\n\ndef agent_loop(messages: list):\n while True:\n response = client.messages.create(\n model=MODEL, system=SYSTEM,\n messages=normalize_messages(messages),\n tools=TOOLS, max_tokens=8000,\n )\n messages.append({\"role\": \"assistant\", \"content\": response.content})\n if response.stop_reason != \"tool_use\":\n return\n results = []\n for block in response.content:\n if block.type == \"tool_use\":\n handler = TOOL_HANDLERS.get(block.name)\n output = handler(**block.input) if handler else f\"Unknown tool: {block.name}\"\n print(f\"> {block.name}: {output[:200]}\")\n results.append({\"type\": \"tool_result\", \"tool_use_id\": block.id, \"content\": output})\n messages.append({\"role\": \"user\", \"content\": results})\n\n\nif __name__ == \"__main__\":\n history = []\n while True:\n try:\n query = input(\"\\033[36ms02 >> \\033[0m\")\n except (EOFError, KeyboardInterrupt):\n break\n if query.strip().lower() in (\"q\", \"exit\", \"\"):\n break\n history.append({\"role\": \"user\", \"content\": query})\n agent_loop(history)\n response_content = history[-1][\"content\"]\n if isinstance(response_content, list):\n for block in response_content:\n if hasattr(block, \"text\"):\n print(block.text)\n print()\n" + "source": "#!/usr/bin/env python3\n# Harness: tool dispatch -- expanding what the model can reach.\n\"\"\"\ns02_tool_use.py - Tool dispatch + message normalization\n\nThe agent loop from s01 didn't change. We added tools to the dispatch map,\nand a normalize_messages() function that cleans up the message list before\neach API call.\n\nKey insight: \"The loop didn't change at all. I just added tools.\"\n\"\"\"\n\nimport os\nimport subprocess\nfrom pathlib import Path\n\nfrom anthropic import Anthropic\nfrom dotenv import load_dotenv\n\nload_dotenv(override=True)\n\nif os.getenv(\"ANTHROPIC_BASE_URL\"):\n os.environ.pop(\"ANTHROPIC_AUTH_TOKEN\", None)\n\nWORKDIR = Path.cwd()\nclient = Anthropic(base_url=os.getenv(\"ANTHROPIC_BASE_URL\"))\nMODEL = os.environ[\"MODEL_ID\"]\n\nSYSTEM = f\"You are a coding agent at {WORKDIR}. Use tools to solve tasks. Act, don't explain.\"\n\n\ndef safe_path(p: str) -> Path:\n path = (WORKDIR / p).resolve()\n if not path.is_relative_to(WORKDIR):\n raise ValueError(f\"Path escapes workspace: {p}\")\n return path\n\n\ndef run_bash(command: str) -> str:\n dangerous = [\"rm -rf /\", \"sudo\", \"shutdown\", \"reboot\", \"> /dev/\"]\n if any(d in command for d in dangerous):\n return \"Error: Dangerous command blocked\"\n try:\n r = subprocess.run(command, shell=True, cwd=WORKDIR,\n capture_output=True, text=True, timeout=120)\n out = (r.stdout + r.stderr).strip()\n return out[:50000] if out else \"(no output)\"\n except subprocess.TimeoutExpired:\n return \"Error: Timeout (120s)\"\n\n\ndef run_read(path: str, limit: int = None) -> str:\n try:\n text = safe_path(path).read_text()\n lines = text.splitlines()\n if limit and limit < len(lines):\n lines = lines[:limit] + [f\"... ({len(lines) - limit} more lines)\"]\n return \"\\n\".join(lines)[:50000]\n except Exception as e:\n return f\"Error: {e}\"\n\n\ndef run_write(path: str, content: str) -> str:\n try:\n fp = safe_path(path)\n fp.parent.mkdir(parents=True, exist_ok=True)\n fp.write_text(content)\n return f\"Wrote {len(content)} bytes to {path}\"\n except Exception as e:\n return f\"Error: {e}\"\n\n\ndef run_edit(path: str, old_text: str, new_text: str) -> str:\n try:\n fp = safe_path(path)\n content = fp.read_text()\n if old_text not in content:\n return f\"Error: Text not found in {path}\"\n fp.write_text(content.replace(old_text, new_text, 1))\n return f\"Edited {path}\"\n except Exception as e:\n return f\"Error: {e}\"\n\n\n# -- Concurrency safety classification --\n# Read-only tools can safely run in parallel; mutating tools must be serialized.\nCONCURRENCY_SAFE = {\"read_file\"}\nCONCURRENCY_UNSAFE = {\"write_file\", \"edit_file\"}\n\n# -- The dispatch map: {tool_name: handler} --\nTOOL_HANDLERS = {\n \"bash\": lambda **kw: run_bash(kw[\"command\"]),\n \"read_file\": lambda **kw: run_read(kw[\"path\"], kw.get(\"limit\")),\n \"write_file\": lambda **kw: run_write(kw[\"path\"], kw[\"content\"]),\n \"edit_file\": lambda **kw: run_edit(kw[\"path\"], kw[\"old_text\"], kw[\"new_text\"]),\n}\n\nTOOLS = [\n {\"name\": \"bash\", \"description\": \"Run a shell command.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"command\": {\"type\": \"string\"}}, \"required\": [\"command\"]}},\n {\"name\": \"read_file\", \"description\": \"Read file contents.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"limit\": {\"type\": \"integer\"}}, \"required\": [\"path\"]}},\n {\"name\": \"write_file\", \"description\": \"Write content to file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"content\": {\"type\": \"string\"}}, \"required\": [\"path\", \"content\"]}},\n {\"name\": \"edit_file\", \"description\": \"Replace exact text in file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"old_text\": {\"type\": \"string\"}, \"new_text\": {\"type\": \"string\"}}, \"required\": [\"path\", \"old_text\", \"new_text\"]}},\n]\n\n\ndef normalize_messages(messages: list) -> list:\n \"\"\"Clean up messages before sending to the API.\n\n Three jobs:\n 1. Strip internal metadata fields the API doesn't understand\n 2. Ensure every tool_use has a matching tool_result (insert placeholder if missing)\n 3. Merge consecutive same-role messages (API requires strict alternation)\n \"\"\"\n cleaned = []\n for msg in messages:\n clean = {\"role\": msg[\"role\"]}\n if isinstance(msg.get(\"content\"), str):\n clean[\"content\"] = msg[\"content\"]\n elif isinstance(msg.get(\"content\"), list):\n clean[\"content\"] = [\n {k: v for k, v in block.items()\n if not k.startswith(\"_\")}\n for block in msg[\"content\"]\n if isinstance(block, dict)\n ]\n else:\n clean[\"content\"] = msg.get(\"content\", \"\")\n cleaned.append(clean)\n\n # Collect existing tool_result IDs\n existing_results = set()\n for msg in cleaned:\n if isinstance(msg.get(\"content\"), list):\n for block in msg[\"content\"]:\n if isinstance(block, dict) and block.get(\"type\") == \"tool_result\":\n existing_results.add(block.get(\"tool_use_id\"))\n\n # Find orphaned tool_use blocks and insert placeholder results\n for msg in cleaned:\n if msg[\"role\"] != \"assistant\" or not isinstance(msg.get(\"content\"), list):\n continue\n for block in msg[\"content\"]:\n if not isinstance(block, dict):\n continue\n if block.get(\"type\") == \"tool_use\" and block.get(\"id\") not in existing_results:\n cleaned.append({\"role\": \"user\", \"content\": [\n {\"type\": \"tool_result\", \"tool_use_id\": block[\"id\"],\n \"content\": \"(cancelled)\"}\n ]})\n\n # Merge consecutive same-role messages\n if not cleaned:\n return cleaned\n merged = [cleaned[0]]\n for msg in cleaned[1:]:\n if msg[\"role\"] == merged[-1][\"role\"]:\n prev = merged[-1]\n prev_c = prev[\"content\"] if isinstance(prev[\"content\"], list) \\\n else [{\"type\": \"text\", \"text\": str(prev[\"content\"])}]\n curr_c = msg[\"content\"] if isinstance(msg[\"content\"], list) \\\n else [{\"type\": \"text\", \"text\": str(msg[\"content\"])}]\n prev[\"content\"] = prev_c + curr_c\n else:\n merged.append(msg)\n return merged\n\n\ndef agent_loop(messages: list):\n while True:\n response = client.messages.create(\n model=MODEL, system=SYSTEM,\n messages=normalize_messages(messages),\n tools=TOOLS, max_tokens=8000,\n )\n messages.append({\"role\": \"assistant\", \"content\": response.content})\n if response.stop_reason != \"tool_use\":\n return\n results = []\n for block in response.content:\n if block.type == \"tool_use\":\n handler = TOOL_HANDLERS.get(block.name)\n output = handler(**block.input) if handler else f\"Unknown tool: {block.name}\"\n print(f\"> {block.name}:\")\n print(output[:200])\n results.append({\"type\": \"tool_result\", \"tool_use_id\": block.id, \"content\": output})\n messages.append({\"role\": \"user\", \"content\": results})\n\n\nif __name__ == \"__main__\":\n history = []\n while True:\n try:\n query = input(\"\\033[36ms02 >> \\033[0m\")\n except (EOFError, KeyboardInterrupt):\n break\n if query.strip().lower() in (\"q\", \"exit\", \"\"):\n break\n history.append({\"role\": \"user\", \"content\": query})\n agent_loop(history)\n response_content = history[-1][\"content\"]\n if isinstance(response_content, list):\n for block in response_content:\n if hasattr(block, \"text\"):\n print(block.text)\n print()\n" }, { "id": "s03", @@ -191,7 +191,7 @@ "filename": "s04_subagent.py", "title": "Subagent", "subtitle": "Fresh Context per Subtask", - "loc": 197, + "loc": 200, "tools": [ "bash", "read_file", @@ -225,31 +225,31 @@ { "name": "run_read", "signature": "def run_read(path: str, limit: int = None)", - "startLine": 117 + "startLine": 119 }, { "name": "run_write", "signature": "def run_write(path: str, content: str)", - "startLine": 126 + "startLine": 128 }, { "name": "run_edit", "signature": "def run_edit(path: str, old_text: str, new_text: str)", - "startLine": 135 + "startLine": 137 }, { "name": "run_subagent", "signature": "def run_subagent(prompt: str)", - "startLine": 168 + "startLine": 170 }, { "name": "agent_loop", "signature": "def agent_loop(messages: list)", - "startLine": 196 + "startLine": 198 } ], "layer": "core", - "source": "#!/usr/bin/env python3\n# Harness: context isolation -- protecting the model's clarity of thought.\n\"\"\"\ns04_subagent.py - Subagents\n\nSpawn a child agent with fresh messages=[]. The child works in its own\ncontext, sharing the filesystem, then returns only a summary to the parent.\n\n Parent agent Subagent\n +------------------+ +------------------+\n | messages=[...] | | messages=[] | <-- fresh\n | | dispatch | |\n | tool: task | ---------->| while tool_use: |\n | prompt=\"...\" | | call tools |\n | description=\"\" | | append results |\n | | summary | |\n | result = \"...\" | <--------- | return last text |\n +------------------+ +------------------+\n |\n Parent context stays clean.\n Subagent context is discarded.\n\nKey insight: \"Fresh messages=[] gives context isolation. The parent stays clean.\"\n\nNote: Real Claude Code also uses in-process isolation (not OS-level process\nforking). The child runs in the same process with a fresh message array and\nisolated tool context -- same pattern as this teaching implementation.\n\n Comparison with real Claude Code:\n +-------------------+------------------+----------------------------------+\n | Aspect | This demo | Real Claude Code |\n +-------------------+------------------+----------------------------------+\n | Backend | in-process only | 5 backends: in-process, tmux, |\n | | | iTerm2, fork, remote |\n | Context isolation | fresh messages=[]| createSubagentContext() isolates |\n | | | ~20 fields (tools, permissions, |\n | | | cwd, env, hooks, etc.) |\n | Tool filtering | manually curated | resolveAgentTools() filters from |\n | | | parent pool; allowedTools |\n | | | replaces all allow rules |\n | Agent definition | hardcoded system | .claude/agents/*.md with YAML |\n | | prompt | frontmatter (AgentTemplate) |\n +-------------------+------------------+----------------------------------+\n\"\"\"\n\nimport os\nimport re\nimport subprocess\nfrom pathlib import Path\n\nfrom anthropic import Anthropic\nfrom dotenv import load_dotenv\n\nload_dotenv(override=True)\n\nif os.getenv(\"ANTHROPIC_BASE_URL\"):\n os.environ.pop(\"ANTHROPIC_AUTH_TOKEN\", None)\n\nWORKDIR = Path.cwd()\nclient = Anthropic(base_url=os.getenv(\"ANTHROPIC_BASE_URL\"))\nMODEL = os.environ[\"MODEL_ID\"]\n\nSYSTEM = f\"You are a coding agent at {WORKDIR}. Use the task tool to delegate exploration or subtasks.\"\nSUBAGENT_SYSTEM = f\"You are a coding subagent at {WORKDIR}. Complete the given task, then summarize your findings.\"\n\n\nclass AgentTemplate:\n \"\"\"\n Parse agent definition from markdown frontmatter.\n\n Real Claude Code loads agent definitions from .claude/agents/*.md.\n Frontmatter fields: name, tools, disallowedTools, skills, hooks,\n model, effort, permissionMode, maxTurns, memory, isolation, color,\n background, initialPrompt, mcpServers.\n 3 sources: built-in, custom (.claude/agents/), plugin-provided.\n \"\"\"\n def __init__(self, path):\n self.path = Path(path)\n self.name = self.path.stem\n self.config = {}\n self.system_prompt = \"\"\n self._parse()\n\n def _parse(self):\n text = self.path.read_text()\n match = re.match(r\"^---\\s*\\n(.*?)\\n---\\s*\\n(.*)\", text, re.DOTALL)\n if not match:\n self.system_prompt = text\n return\n for line in match.group(1).splitlines():\n if \":\" in line:\n k, _, v = line.partition(\":\")\n self.config[k.strip()] = v.strip()\n self.system_prompt = match.group(2).strip()\n self.name = self.config.get(\"name\", self.name)\n\n\n# -- Tool implementations shared by parent and child --\ndef safe_path(p: str) -> Path:\n path = (WORKDIR / p).resolve()\n if not path.is_relative_to(WORKDIR):\n raise ValueError(f\"Path escapes workspace: {p}\")\n return path\n\ndef run_bash(command: str) -> str:\n dangerous = [\"rm -rf /\", \"sudo\", \"shutdown\", \"reboot\", \"> /dev/\"]\n if any(d in command for d in dangerous):\n return \"Error: Dangerous command blocked\"\n try:\n r = subprocess.run(command, shell=True, cwd=WORKDIR,\n capture_output=True, text=True, timeout=120)\n out = (r.stdout + r.stderr).strip()\n return out[:50000] if out else \"(no output)\"\n except subprocess.TimeoutExpired:\n return \"Error: Timeout (120s)\"\n\ndef run_read(path: str, limit: int = None) -> str:\n try:\n lines = safe_path(path).read_text().splitlines()\n if limit and limit < len(lines):\n lines = lines[:limit] + [f\"... ({len(lines) - limit} more)\"]\n return \"\\n\".join(lines)[:50000]\n except Exception as e:\n return f\"Error: {e}\"\n\ndef run_write(path: str, content: str) -> str:\n try:\n fp = safe_path(path)\n fp.parent.mkdir(parents=True, exist_ok=True)\n fp.write_text(content)\n return f\"Wrote {len(content)} bytes\"\n except Exception as e:\n return f\"Error: {e}\"\n\ndef run_edit(path: str, old_text: str, new_text: str) -> str:\n try:\n fp = safe_path(path)\n content = fp.read_text()\n if old_text not in content:\n return f\"Error: Text not found in {path}\"\n fp.write_text(content.replace(old_text, new_text, 1))\n return f\"Edited {path}\"\n except Exception as e:\n return f\"Error: {e}\"\n\n\nTOOL_HANDLERS = {\n \"bash\": lambda **kw: run_bash(kw[\"command\"]),\n \"read_file\": lambda **kw: run_read(kw[\"path\"], kw.get(\"limit\")),\n \"write_file\": lambda **kw: run_write(kw[\"path\"], kw[\"content\"]),\n \"edit_file\": lambda **kw: run_edit(kw[\"path\"], kw[\"old_text\"], kw[\"new_text\"]),\n}\n\n# Child gets all base tools except task (no recursive spawning)\nCHILD_TOOLS = [\n {\"name\": \"bash\", \"description\": \"Run a shell command.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"command\": {\"type\": \"string\"}}, \"required\": [\"command\"]}},\n {\"name\": \"read_file\", \"description\": \"Read file contents.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"limit\": {\"type\": \"integer\"}}, \"required\": [\"path\"]}},\n {\"name\": \"write_file\", \"description\": \"Write content to file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"content\": {\"type\": \"string\"}}, \"required\": [\"path\", \"content\"]}},\n {\"name\": \"edit_file\", \"description\": \"Replace exact text in file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"old_text\": {\"type\": \"string\"}, \"new_text\": {\"type\": \"string\"}}, \"required\": [\"path\", \"old_text\", \"new_text\"]}},\n]\n\n\n# -- Subagent: fresh context, filtered tools, summary-only return --\ndef run_subagent(prompt: str) -> str:\n sub_messages = [{\"role\": \"user\", \"content\": prompt}] # fresh context\n for _ in range(30): # safety limit\n response = client.messages.create(\n model=MODEL, system=SUBAGENT_SYSTEM, messages=sub_messages,\n tools=CHILD_TOOLS, max_tokens=8000,\n )\n sub_messages.append({\"role\": \"assistant\", \"content\": response.content})\n if response.stop_reason != \"tool_use\":\n break\n results = []\n for block in response.content:\n if block.type == \"tool_use\":\n handler = TOOL_HANDLERS.get(block.name)\n output = handler(**block.input) if handler else f\"Unknown tool: {block.name}\"\n results.append({\"type\": \"tool_result\", \"tool_use_id\": block.id, \"content\": str(output)[:50000]})\n sub_messages.append({\"role\": \"user\", \"content\": results})\n # Only the final text returns to the parent -- child context is discarded\n return \"\".join(b.text for b in response.content if hasattr(b, \"text\")) or \"(no summary)\"\n\n\n# -- Parent tools: base tools + task dispatcher --\nPARENT_TOOLS = CHILD_TOOLS + [\n {\"name\": \"task\", \"description\": \"Spawn a subagent with fresh context. It shares the filesystem but not conversation history.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"prompt\": {\"type\": \"string\"}, \"description\": {\"type\": \"string\", \"description\": \"Short description of the task\"}}, \"required\": [\"prompt\"]}},\n]\n\n\ndef agent_loop(messages: list):\n while True:\n response = client.messages.create(\n model=MODEL, system=SYSTEM, messages=messages,\n tools=PARENT_TOOLS, max_tokens=8000,\n )\n messages.append({\"role\": \"assistant\", \"content\": response.content})\n if response.stop_reason != \"tool_use\":\n return\n results = []\n for block in response.content:\n if block.type == \"tool_use\":\n if block.name == \"task\":\n desc = block.input.get(\"description\", \"subtask\")\n print(f\"> task ({desc}): {block.input['prompt'][:80]}\")\n output = run_subagent(block.input[\"prompt\"])\n else:\n handler = TOOL_HANDLERS.get(block.name)\n output = handler(**block.input) if handler else f\"Unknown tool: {block.name}\"\n print(f\" {str(output)[:200]}\")\n results.append({\"type\": \"tool_result\", \"tool_use_id\": block.id, \"content\": str(output)})\n messages.append({\"role\": \"user\", \"content\": results})\n\n\nif __name__ == \"__main__\":\n history = []\n while True:\n try:\n query = input(\"\\033[36ms04 >> \\033[0m\")\n except (EOFError, KeyboardInterrupt):\n break\n if query.strip().lower() in (\"q\", \"exit\", \"\"):\n break\n history.append({\"role\": \"user\", \"content\": query})\n agent_loop(history)\n response_content = history[-1][\"content\"]\n if isinstance(response_content, list):\n for block in response_content:\n if hasattr(block, \"text\"):\n print(block.text)\n print()\n" + "source": "#!/usr/bin/env python3\n# Harness: context isolation -- protecting the model's clarity of thought.\n\"\"\"\ns04_subagent.py - Subagents\n\nSpawn a child agent with fresh messages=[]. The child works in its own\ncontext, sharing the filesystem, then returns only a summary to the parent.\n\n Parent agent Subagent\n +------------------+ +------------------+\n | messages=[...] | | messages=[] | <-- fresh\n | | dispatch | |\n | tool: task | ---------->| while tool_use: |\n | prompt=\"...\" | | call tools |\n | description=\"\" | | append results |\n | | summary | |\n | result = \"...\" | <--------- | return last text |\n +------------------+ +------------------+\n |\n Parent context stays clean.\n Subagent context is discarded.\n\nKey insight: \"Fresh messages=[] gives context isolation. The parent stays clean.\"\n\nNote: Real Claude Code also uses in-process isolation (not OS-level process\nforking). The child runs in the same process with a fresh message array and\nisolated tool context -- same pattern as this teaching implementation.\n\n Comparison with real Claude Code:\n +-------------------+------------------+----------------------------------+\n | Aspect | This demo | Real Claude Code |\n +-------------------+------------------+----------------------------------+\n | Backend | in-process only | 5 backends: in-process, tmux, |\n | | | iTerm2, fork, remote |\n | Context isolation | fresh messages=[]| createSubagentContext() isolates |\n | | | ~20 fields (tools, permissions, |\n | | | cwd, env, hooks, etc.) |\n | Tool filtering | manually curated | resolveAgentTools() filters from |\n | | | parent pool; allowedTools |\n | | | replaces all allow rules |\n | Agent definition | hardcoded system | .claude/agents/*.md with YAML |\n | | prompt | frontmatter (AgentTemplate) |\n +-------------------+------------------+----------------------------------+\n\"\"\"\n\nimport os\nimport re\nimport subprocess\nfrom pathlib import Path\n\nfrom anthropic import Anthropic\nfrom dotenv import load_dotenv\n\nload_dotenv(override=True)\n\nif os.getenv(\"ANTHROPIC_BASE_URL\"):\n os.environ.pop(\"ANTHROPIC_AUTH_TOKEN\", None)\n\nWORKDIR = Path.cwd()\nclient = Anthropic(base_url=os.getenv(\"ANTHROPIC_BASE_URL\"))\nMODEL = os.environ[\"MODEL_ID\"]\n\nSYSTEM = f\"You are a coding agent at {WORKDIR}. Use the task tool to delegate exploration or subtasks.\"\nSUBAGENT_SYSTEM = f\"You are a coding subagent at {WORKDIR}. Complete the given task, then summarize your findings.\"\n\n\nclass AgentTemplate:\n \"\"\"\n Parse agent definition from markdown frontmatter.\n\n Real Claude Code loads agent definitions from .claude/agents/*.md.\n Frontmatter fields: name, tools, disallowedTools, skills, hooks,\n model, effort, permissionMode, maxTurns, memory, isolation, color,\n background, initialPrompt, mcpServers.\n 3 sources: built-in, custom (.claude/agents/), plugin-provided.\n \"\"\"\n def __init__(self, path):\n self.path = Path(path)\n self.name = self.path.stem\n self.config = {}\n self.system_prompt = \"\"\n self._parse()\n\n def _parse(self):\n text = self.path.read_text()\n match = re.match(r\"^---\\s*\\n(.*?)\\n---\\s*\\n(.*)\", text, re.DOTALL)\n if not match:\n self.system_prompt = text\n return\n for line in match.group(1).splitlines():\n if \":\" in line:\n k, _, v = line.partition(\":\")\n self.config[k.strip()] = v.strip()\n self.system_prompt = match.group(2).strip()\n self.name = self.config.get(\"name\", self.name)\n\n\n# -- Tool implementations shared by parent and child --\ndef safe_path(p: str) -> Path:\n path = (WORKDIR / p).resolve()\n if not path.is_relative_to(WORKDIR):\n raise ValueError(f\"Path escapes workspace: {p}\")\n return path\n\ndef run_bash(command: str) -> str:\n dangerous = [\"rm -rf /\", \"sudo\", \"shutdown\", \"reboot\", \"> /dev/\"]\n if any(d in command for d in dangerous):\n return \"Error: Dangerous command blocked\"\n try:\n r = subprocess.run(command, shell=True, cwd=WORKDIR,\n capture_output=True, text=True, timeout=120)\n out = (r.stdout + r.stderr).strip()\n return out[:50000] if out else \"(no output)\"\n except subprocess.TimeoutExpired:\n return \"Error: Timeout (120s)\"\n except (FileNotFoundError, OSError) as e:\n return f\"Error: {e}\"\n\ndef run_read(path: str, limit: int = None) -> str:\n try:\n lines = safe_path(path).read_text().splitlines()\n if limit and limit < len(lines):\n lines = lines[:limit] + [f\"... ({len(lines) - limit} more)\"]\n return \"\\n\".join(lines)[:50000]\n except Exception as e:\n return f\"Error: {e}\"\n\ndef run_write(path: str, content: str) -> str:\n try:\n fp = safe_path(path)\n fp.parent.mkdir(parents=True, exist_ok=True)\n fp.write_text(content)\n return f\"Wrote {len(content)} bytes\"\n except Exception as e:\n return f\"Error: {e}\"\n\ndef run_edit(path: str, old_text: str, new_text: str) -> str:\n try:\n fp = safe_path(path)\n content = fp.read_text()\n if old_text not in content:\n return f\"Error: Text not found in {path}\"\n fp.write_text(content.replace(old_text, new_text, 1))\n return f\"Edited {path}\"\n except Exception as e:\n return f\"Error: {e}\"\n\n\nTOOL_HANDLERS = {\n \"bash\": lambda **kw: run_bash(kw[\"command\"]),\n \"read_file\": lambda **kw: run_read(kw[\"path\"], kw.get(\"limit\")),\n \"write_file\": lambda **kw: run_write(kw[\"path\"], kw[\"content\"]),\n \"edit_file\": lambda **kw: run_edit(kw[\"path\"], kw[\"old_text\"], kw[\"new_text\"]),\n}\n\n# Child gets all base tools except task (no recursive spawning)\nCHILD_TOOLS = [\n {\"name\": \"bash\", \"description\": \"Run a shell command.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"command\": {\"type\": \"string\"}}, \"required\": [\"command\"]}},\n {\"name\": \"read_file\", \"description\": \"Read file contents.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"limit\": {\"type\": \"integer\"}}, \"required\": [\"path\"]}},\n {\"name\": \"write_file\", \"description\": \"Write content to file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"content\": {\"type\": \"string\"}}, \"required\": [\"path\", \"content\"]}},\n {\"name\": \"edit_file\", \"description\": \"Replace exact text in file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"old_text\": {\"type\": \"string\"}, \"new_text\": {\"type\": \"string\"}}, \"required\": [\"path\", \"old_text\", \"new_text\"]}},\n]\n\n\n# -- Subagent: fresh context, filtered tools, summary-only return --\ndef run_subagent(prompt: str) -> str:\n sub_messages = [{\"role\": \"user\", \"content\": prompt}] # fresh context\n for _ in range(30): # safety limit\n response = client.messages.create(\n model=MODEL, system=SUBAGENT_SYSTEM, messages=sub_messages,\n tools=CHILD_TOOLS, max_tokens=8000,\n )\n sub_messages.append({\"role\": \"assistant\", \"content\": response.content})\n if response.stop_reason != \"tool_use\":\n break\n results = []\n for block in response.content:\n if block.type == \"tool_use\":\n handler = TOOL_HANDLERS.get(block.name)\n output = handler(**block.input) if handler else f\"Unknown tool: {block.name}\"\n results.append({\"type\": \"tool_result\", \"tool_use_id\": block.id, \"content\": str(output)[:50000]})\n sub_messages.append({\"role\": \"user\", \"content\": results})\n # Only the final text returns to the parent -- child context is discarded\n return \"\".join(b.text for b in response.content if hasattr(b, \"text\")) or \"(no summary)\"\n\n\n# -- Parent tools: base tools + task dispatcher --\nPARENT_TOOLS = CHILD_TOOLS + [\n {\"name\": \"task\", \"description\": \"Spawn a subagent with fresh context. It shares the filesystem but not conversation history.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"prompt\": {\"type\": \"string\"}, \"description\": {\"type\": \"string\", \"description\": \"Short description of the task\"}}, \"required\": [\"prompt\"]}},\n]\n\n\ndef agent_loop(messages: list):\n while True:\n response = client.messages.create(\n model=MODEL, system=SYSTEM, messages=messages,\n tools=PARENT_TOOLS, max_tokens=8000,\n )\n messages.append({\"role\": \"assistant\", \"content\": response.content})\n if response.stop_reason != \"tool_use\":\n return\n results = []\n for block in response.content:\n if block.type == \"tool_use\":\n if block.name == \"task\":\n desc = block.input.get(\"description\", \"subtask\")\n prompt = block.input.get(\"prompt\", \"\")\n print(f\"> task ({desc}): {prompt[:80]}\")\n output = run_subagent(prompt)\n else:\n handler = TOOL_HANDLERS.get(block.name)\n output = handler(**block.input) if handler else f\"Unknown tool: {block.name}\"\n print(f\" {str(output)[:200]}\")\n results.append({\"type\": \"tool_result\", \"tool_use_id\": block.id, \"content\": str(output)})\n messages.append({\"role\": \"user\", \"content\": results})\n\n\nif __name__ == \"__main__\":\n history = []\n while True:\n try:\n query = input(\"\\033[36ms04 >> \\033[0m\")\n except (EOFError, KeyboardInterrupt):\n break\n if query.strip().lower() in (\"q\", \"exit\", \"\"):\n break\n history.append({\"role\": \"user\", \"content\": query})\n agent_loop(history)\n response_content = history[-1][\"content\"]\n if isinstance(response_content, list):\n for block in response_content:\n if hasattr(block, \"text\"):\n print(block.text)\n print()\n" }, { "id": "s05", @@ -894,7 +894,7 @@ } ], "layer": "runtime", - "source": "#!/usr/bin/env python3\n# Harness: background execution -- the model thinks while the harness waits.\n\"\"\"\ns13_background_tasks.py - Background Tasks\n\nRun slow commands in background threads. Before each LLM call, the loop\ndrains a notification queue and hands finished results back to the model.\n\n Main thread Background thread\n +-----------------+ +-----------------+\n | agent loop | | task executes |\n | ... | | ... |\n | [LLM call] <---+------- | enqueue(result) |\n | ^drain queue | +-----------------+\n +-----------------+\n\n Timeline:\n Agent ----[spawn A]----[spawn B]----[other work]----\n | |\n v v\n [A runs] [B runs]\n | |\n +-- notification queue --> [results injected]\n\nBackground tasks here are runtime execution slots, not the durable task-board\nrecords introduced in s12.\n\"\"\"\n\nimport os\nimport json\nimport subprocess\nimport threading\nimport time\nimport uuid\nfrom pathlib import Path\n\nfrom anthropic import Anthropic\nfrom dotenv import load_dotenv\n\nload_dotenv(override=True)\n\nif os.getenv(\"ANTHROPIC_BASE_URL\"):\n os.environ.pop(\"ANTHROPIC_AUTH_TOKEN\", None)\n\nWORKDIR = Path.cwd()\nRUNTIME_DIR = WORKDIR / \".runtime-tasks\"\nRUNTIME_DIR.mkdir(exist_ok=True)\nclient = Anthropic(base_url=os.getenv(\"ANTHROPIC_BASE_URL\"))\nMODEL = os.environ[\"MODEL_ID\"]\n\nSYSTEM = f\"You are a coding agent at {WORKDIR}. Use background_run for long-running commands.\"\n\nSTALL_THRESHOLD_S = 45 # seconds before a task is considered stalled\n\n\nclass NotificationQueue:\n \"\"\"\n Priority-based notification queue with same-key folding.\n\n Folding means a newer message can replace an older message with the\n same key, so the context is not flooded with stale updates.\n \"\"\"\n\n PRIORITIES = {\"immediate\": 0, \"high\": 1, \"medium\": 2, \"low\": 3}\n\n def __init__(self):\n self._queue = [] # list of (priority, key, message)\n self._lock = threading.Lock()\n\n def push(self, message: str, priority: str = \"medium\", key: str = None):\n \"\"\"Add a message to the queue, folding if key matches an existing entry.\"\"\"\n with self._lock:\n if key:\n # Fold: replace existing message with same key\n self._queue = [(p, k, m) for p, k, m in self._queue if k != key]\n self._queue.append((self.PRIORITIES.get(priority, 2), key, message))\n self._queue.sort(key=lambda x: x[0])\n\n def drain(self) -> list[str]:\n \"\"\"Return all pending messages in priority order and clear the queue.\"\"\"\n with self._lock:\n messages = [m for _, _, m in self._queue]\n self._queue.clear()\n return messages\n\n\n# -- BackgroundManager: threaded execution + notification queue --\nclass BackgroundManager:\n def __init__(self):\n self.dir = RUNTIME_DIR\n self.tasks = {} # task_id -> {status, result, command, started_at}\n self._notification_queue = [] # completed task results\n self._lock = threading.Lock()\n\n def _record_path(self, task_id: str) -> Path:\n return self.dir / f\"{task_id}.json\"\n\n def _output_path(self, task_id: str) -> Path:\n return self.dir / f\"{task_id}.log\"\n\n def _persist_task(self, task_id: str):\n record = dict(self.tasks[task_id])\n self._record_path(task_id).write_text(\n json.dumps(record, indent=2, ensure_ascii=False)\n )\n\n def _preview(self, output: str, limit: int = 500) -> str:\n compact = \" \".join((output or \"(no output)\").split())\n return compact[:limit]\n\n def run(self, command: str) -> str:\n \"\"\"Start a background thread, return task_id immediately.\"\"\"\n task_id = str(uuid.uuid4())[:8]\n output_file = self._output_path(task_id)\n self.tasks[task_id] = {\n \"id\": task_id,\n \"status\": \"running\",\n \"result\": None,\n \"command\": command,\n \"started_at\": time.time(),\n \"finished_at\": None,\n \"result_preview\": \"\",\n \"output_file\": str(output_file.relative_to(WORKDIR)),\n }\n self._persist_task(task_id)\n thread = threading.Thread(\n target=self._execute, args=(task_id, command), daemon=True\n )\n thread.start()\n return (\n f\"Background task {task_id} started: {command[:80]} \"\n f\"(output_file={output_file.relative_to(WORKDIR)})\"\n )\n\n def _execute(self, task_id: str, command: str):\n \"\"\"Thread target: run subprocess, capture output, push to queue.\"\"\"\n try:\n r = subprocess.run(\n command, shell=True, cwd=WORKDIR,\n capture_output=True, text=True, timeout=300\n )\n output = (r.stdout + r.stderr).strip()[:50000]\n status = \"completed\"\n except subprocess.TimeoutExpired:\n output = \"Error: Timeout (300s)\"\n status = \"timeout\"\n except Exception as e:\n output = f\"Error: {e}\"\n status = \"error\"\n final_output = output or \"(no output)\"\n preview = self._preview(final_output)\n output_path = self._output_path(task_id)\n output_path.write_text(final_output)\n self.tasks[task_id][\"status\"] = status\n self.tasks[task_id][\"result\"] = final_output\n self.tasks[task_id][\"finished_at\"] = time.time()\n self.tasks[task_id][\"result_preview\"] = preview\n self._persist_task(task_id)\n with self._lock:\n self._notification_queue.append({\n \"task_id\": task_id,\n \"status\": status,\n \"command\": command[:80],\n \"preview\": preview,\n \"output_file\": str(output_path.relative_to(WORKDIR)),\n })\n\n def check(self, task_id: str = None) -> str:\n \"\"\"Check status of one task or list all.\"\"\"\n if task_id:\n t = self.tasks.get(task_id)\n if not t:\n return f\"Error: Unknown task {task_id}\"\n visible = {\n \"id\": t[\"id\"],\n \"status\": t[\"status\"],\n \"command\": t[\"command\"],\n \"result_preview\": t.get(\"result_preview\", \"\"),\n \"output_file\": t.get(\"output_file\", \"\"),\n }\n return json.dumps(visible, indent=2, ensure_ascii=False)\n lines = []\n for tid, t in self.tasks.items():\n lines.append(\n f\"{tid}: [{t['status']}] {t['command'][:60]} \"\n f\"-> {t.get('result_preview') or '(running)'}\"\n )\n return \"\\n\".join(lines) if lines else \"No background tasks.\"\n\n def drain_notifications(self) -> list:\n \"\"\"Return and clear all pending completion notifications.\"\"\"\n with self._lock:\n notifs = list(self._notification_queue)\n self._notification_queue.clear()\n return notifs\n\n def detect_stalled(self) -> list[str]:\n \"\"\"\n Return task IDs that have been running longer than STALL_THRESHOLD_S.\n \"\"\"\n now = time.time()\n stalled = []\n for task_id, info in self.tasks.items():\n if info[\"status\"] != \"running\":\n continue\n elapsed = now - info.get(\"started_at\", now)\n if elapsed > STALL_THRESHOLD_S:\n stalled.append(task_id)\n return stalled\n\n\nBG = BackgroundManager()\n\n\n# -- Tool implementations --\ndef safe_path(p: str) -> Path:\n path = (WORKDIR / p).resolve()\n if not path.is_relative_to(WORKDIR):\n raise ValueError(f\"Path escapes workspace: {p}\")\n return path\n\ndef run_bash(command: str) -> str:\n dangerous = [\"rm -rf /\", \"sudo\", \"shutdown\", \"reboot\", \"> /dev/\"]\n if any(d in command for d in dangerous):\n return \"Error: Dangerous command blocked\"\n try:\n r = subprocess.run(command, shell=True, cwd=WORKDIR,\n capture_output=True, text=True, timeout=120)\n out = (r.stdout + r.stderr).strip()\n return out[:50000] if out else \"(no output)\"\n except subprocess.TimeoutExpired:\n return \"Error: Timeout (120s)\"\n\ndef run_read(path: str, limit: int = None) -> str:\n try:\n lines = safe_path(path).read_text().splitlines()\n if limit and limit < len(lines):\n lines = lines[:limit] + [f\"... ({len(lines) - limit} more)\"]\n return \"\\n\".join(lines)[:50000]\n except Exception as e:\n return f\"Error: {e}\"\n\ndef run_write(path: str, content: str) -> str:\n try:\n fp = safe_path(path)\n fp.parent.mkdir(parents=True, exist_ok=True)\n fp.write_text(content)\n return f\"Wrote {len(content)} bytes\"\n except Exception as e:\n return f\"Error: {e}\"\n\ndef run_edit(path: str, old_text: str, new_text: str) -> str:\n try:\n fp = safe_path(path)\n c = fp.read_text()\n if old_text not in c:\n return f\"Error: Text not found in {path}\"\n fp.write_text(c.replace(old_text, new_text, 1))\n return f\"Edited {path}\"\n except Exception as e:\n return f\"Error: {e}\"\n\n\nTOOL_HANDLERS = {\n \"bash\": lambda **kw: run_bash(kw[\"command\"]),\n \"read_file\": lambda **kw: run_read(kw[\"path\"], kw.get(\"limit\")),\n \"write_file\": lambda **kw: run_write(kw[\"path\"], kw[\"content\"]),\n \"edit_file\": lambda **kw: run_edit(kw[\"path\"], kw[\"old_text\"], kw[\"new_text\"]),\n \"background_run\": lambda **kw: BG.run(kw[\"command\"]),\n \"check_background\": lambda **kw: BG.check(kw.get(\"task_id\")),\n}\n\nTOOLS = [\n {\"name\": \"bash\", \"description\": \"Run a shell command (blocking).\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"command\": {\"type\": \"string\"}}, \"required\": [\"command\"]}},\n {\"name\": \"read_file\", \"description\": \"Read file contents.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"limit\": {\"type\": \"integer\"}}, \"required\": [\"path\"]}},\n {\"name\": \"write_file\", \"description\": \"Write content to file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"content\": {\"type\": \"string\"}}, \"required\": [\"path\", \"content\"]}},\n {\"name\": \"edit_file\", \"description\": \"Replace exact text in file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"old_text\": {\"type\": \"string\"}, \"new_text\": {\"type\": \"string\"}}, \"required\": [\"path\", \"old_text\", \"new_text\"]}},\n {\"name\": \"background_run\", \"description\": \"Run command in background thread. Returns task_id immediately.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"command\": {\"type\": \"string\"}}, \"required\": [\"command\"]}},\n {\"name\": \"check_background\", \"description\": \"Check background task status. Omit task_id to list all.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"task_id\": {\"type\": \"string\"}}}},\n]\n\n\ndef agent_loop(messages: list):\n while True:\n # Drain background notifications and inject as a synthetic user/assistant\n # transcript pair before the next model call (teaching demo behavior).\n notifs = BG.drain_notifications()\n if notifs and messages:\n notif_text = \"\\n\".join(\n f\"[bg:{n['task_id']}] {n['status']}: {n['preview']} \"\n f\"(output_file={n['output_file']})\"\n for n in notifs\n )\n messages.append({\"role\": \"user\", \"content\": f\"\\n{notif_text}\\n\"})\n messages.append({\"role\": \"assistant\", \"content\": \"Noted background results.\"})\n response = client.messages.create(\n model=MODEL, system=SYSTEM, messages=messages,\n tools=TOOLS, max_tokens=8000,\n )\n messages.append({\"role\": \"assistant\", \"content\": response.content})\n if response.stop_reason != \"tool_use\":\n return\n results = []\n for block in response.content:\n if block.type == \"tool_use\":\n handler = TOOL_HANDLERS.get(block.name)\n try:\n output = handler(**block.input) if handler else f\"Unknown tool: {block.name}\"\n except Exception as e:\n output = f\"Error: {e}\"\n print(f\"> {block.name}: {str(output)[:200]}\")\n results.append({\"type\": \"tool_result\", \"tool_use_id\": block.id, \"content\": str(output)})\n messages.append({\"role\": \"user\", \"content\": results})\n\n\nif __name__ == \"__main__\":\n history = []\n while True:\n try:\n query = input(\"\\033[36ms13 >> \\033[0m\")\n except (EOFError, KeyboardInterrupt):\n break\n if query.strip().lower() in (\"q\", \"exit\", \"\"):\n break\n history.append({\"role\": \"user\", \"content\": query})\n agent_loop(history)\n response_content = history[-1][\"content\"]\n if isinstance(response_content, list):\n for block in response_content:\n if hasattr(block, \"text\"):\n print(block.text)\n print()\n" + "source": "#!/usr/bin/env python3\n# Harness: background execution -- the model thinks while the harness waits.\n\"\"\"\ns13_background_tasks.py - Background Tasks\n\nRun slow commands in background threads. Before each LLM call, the loop\ndrains a notification queue and hands finished results back to the model.\n\n Main thread Background thread\n +-----------------+ +-----------------+\n | agent loop | | task executes |\n | ... | | ... |\n | [LLM call] <---+------- | enqueue(result) |\n | ^drain queue | +-----------------+\n +-----------------+\n\n Timeline:\n Agent ----[spawn A]----[spawn B]----[other work]----\n | |\n v v\n [A runs] [B runs]\n | |\n +-- notification queue --> [results injected]\n\nBackground tasks here are runtime execution slots, not the durable task-board\nrecords introduced in s12.\n\"\"\"\n\nimport os\nimport json\nimport subprocess\nimport threading\nimport time\nimport uuid\nfrom pathlib import Path\n\nfrom anthropic import Anthropic\nfrom dotenv import load_dotenv\n\nload_dotenv(override=True)\n\nif os.getenv(\"ANTHROPIC_BASE_URL\"):\n os.environ.pop(\"ANTHROPIC_AUTH_TOKEN\", None)\n\nWORKDIR = Path.cwd()\nRUNTIME_DIR = WORKDIR / \".runtime-tasks\"\nRUNTIME_DIR.mkdir(exist_ok=True)\nclient = Anthropic(base_url=os.getenv(\"ANTHROPIC_BASE_URL\"))\nMODEL = os.environ[\"MODEL_ID\"]\n\nSYSTEM = f\"You are a coding agent at {WORKDIR}. Use background_run for long-running commands.\"\n\nSTALL_THRESHOLD_S = 45 # seconds before a task is considered stalled\n\n\nclass NotificationQueue:\n \"\"\"\n Priority-based notification queue with same-key folding.\n\n Folding means a newer message can replace an older message with the\n same key, so the context is not flooded with stale updates.\n \"\"\"\n\n PRIORITIES = {\"immediate\": 0, \"high\": 1, \"medium\": 2, \"low\": 3}\n\n def __init__(self):\n self._queue = [] # list of (priority, key, message)\n self._lock = threading.Lock()\n\n def push(self, message: str, priority: str = \"medium\", key: str = None):\n \"\"\"Add a message to the queue, folding if key matches an existing entry.\"\"\"\n with self._lock:\n if key:\n # Fold: replace existing message with same key\n self._queue = [(p, k, m) for p, k, m in self._queue if k != key]\n self._queue.append((self.PRIORITIES.get(priority, 2), key, message))\n self._queue.sort(key=lambda x: x[0])\n\n def drain(self) -> list[str]:\n \"\"\"Return all pending messages in priority order and clear the queue.\"\"\"\n with self._lock:\n messages = [m for _, _, m in self._queue]\n self._queue.clear()\n return messages\n\n\n# -- BackgroundManager: threaded execution + notification queue --\nclass BackgroundManager:\n def __init__(self):\n self.dir = RUNTIME_DIR\n self.tasks = {} # task_id -> {status, result, command, started_at}\n self._notification_queue = [] # completed task results\n self._lock = threading.Lock()\n\n def _record_path(self, task_id: str) -> Path:\n return self.dir / f\"{task_id}.json\"\n\n def _output_path(self, task_id: str) -> Path:\n return self.dir / f\"{task_id}.log\"\n\n def _persist_task(self, task_id: str):\n record = dict(self.tasks[task_id])\n self._record_path(task_id).write_text(\n json.dumps(record, indent=2, ensure_ascii=False)\n )\n\n def _preview(self, output: str, limit: int = 500) -> str:\n compact = \" \".join((output or \"(no output)\").split())\n return compact[:limit]\n\n def run(self, command: str) -> str:\n \"\"\"Start a background thread, return task_id immediately.\"\"\"\n task_id = str(uuid.uuid4())[:8]\n output_file = self._output_path(task_id)\n self.tasks[task_id] = {\n \"id\": task_id,\n \"status\": \"running\",\n \"result\": None,\n \"command\": command,\n \"started_at\": time.time(),\n \"finished_at\": None,\n \"result_preview\": \"\",\n \"output_file\": str(output_file.relative_to(WORKDIR)),\n }\n self._persist_task(task_id)\n thread = threading.Thread(\n target=self._execute, args=(task_id, command), daemon=True\n )\n thread.start()\n return (\n f\"Background task {task_id} started: {command[:80]} \"\n f\"(output_file={output_file.relative_to(WORKDIR)})\"\n )\n\n def _execute(self, task_id: str, command: str):\n \"\"\"Thread target: run subprocess, capture output, push to queue.\"\"\"\n try:\n r = subprocess.run(\n command, shell=True, cwd=WORKDIR,\n capture_output=True, text=True, timeout=300\n )\n output = (r.stdout + r.stderr).strip()[:50000]\n status = \"completed\"\n except subprocess.TimeoutExpired:\n output = \"Error: Timeout (300s)\"\n status = \"timeout\"\n except Exception as e:\n output = f\"Error: {e}\"\n status = \"error\"\n final_output = output or \"(no output)\"\n preview = self._preview(final_output)\n output_path = self._output_path(task_id)\n output_path.write_text(final_output)\n self.tasks[task_id][\"status\"] = status\n self.tasks[task_id][\"result\"] = final_output\n self.tasks[task_id][\"finished_at\"] = time.time()\n self.tasks[task_id][\"result_preview\"] = preview\n self._persist_task(task_id)\n with self._lock:\n self._notification_queue.append({\n \"task_id\": task_id,\n \"status\": status,\n \"command\": command[:80],\n \"preview\": preview,\n \"output_file\": str(output_path.relative_to(WORKDIR)),\n })\n\n def check(self, task_id: str = None) -> str:\n \"\"\"Check status of one task or list all.\"\"\"\n if task_id:\n t = self.tasks.get(task_id)\n if not t:\n return f\"Error: Unknown task {task_id}\"\n visible = {\n \"id\": t[\"id\"],\n \"status\": t[\"status\"],\n \"command\": t[\"command\"],\n \"result_preview\": t.get(\"result_preview\", \"\"),\n \"output_file\": t.get(\"output_file\", \"\"),\n }\n return json.dumps(visible, indent=2, ensure_ascii=False)\n lines = []\n for tid, t in self.tasks.items():\n lines.append(\n f\"{tid}: [{t['status']}] {t['command'][:60]} \"\n f\"-> {t.get('result_preview') or '(running)'}\"\n )\n return \"\\n\".join(lines) if lines else \"No background tasks.\"\n\n def drain_notifications(self) -> list:\n \"\"\"Return and clear all pending completion notifications.\"\"\"\n with self._lock:\n notifs = list(self._notification_queue)\n self._notification_queue.clear()\n return notifs\n\n def detect_stalled(self) -> list[str]:\n \"\"\"\n Return task IDs that have been running longer than STALL_THRESHOLD_S.\n \"\"\"\n now = time.time()\n stalled = []\n for task_id, info in self.tasks.items():\n if info[\"status\"] != \"running\":\n continue\n elapsed = now - info.get(\"started_at\", now)\n if elapsed > STALL_THRESHOLD_S:\n stalled.append(task_id)\n return stalled\n\n\nBG = BackgroundManager()\n\n\n# -- Tool implementations --\ndef safe_path(p: str) -> Path:\n path = (WORKDIR / p).resolve()\n if not path.is_relative_to(WORKDIR):\n raise ValueError(f\"Path escapes workspace: {p}\")\n return path\n\ndef run_bash(command: str) -> str:\n dangerous = [\"rm -rf /\", \"sudo\", \"shutdown\", \"reboot\", \"> /dev/\"]\n if any(d in command for d in dangerous):\n return \"Error: Dangerous command blocked\"\n try:\n r = subprocess.run(command, shell=True, cwd=WORKDIR,\n capture_output=True, text=True, timeout=120)\n out = (r.stdout + r.stderr).strip()\n return out[:50000] if out else \"(no output)\"\n except subprocess.TimeoutExpired:\n return \"Error: Timeout (120s)\"\n\ndef run_read(path: str, limit: int = None) -> str:\n try:\n lines = safe_path(path).read_text().splitlines()\n if limit and limit < len(lines):\n lines = lines[:limit] + [f\"... ({len(lines) - limit} more)\"]\n return \"\\n\".join(lines)[:50000]\n except Exception as e:\n return f\"Error: {e}\"\n\ndef run_write(path: str, content: str) -> str:\n try:\n fp = safe_path(path)\n fp.parent.mkdir(parents=True, exist_ok=True)\n fp.write_text(content)\n return f\"Wrote {len(content)} bytes\"\n except Exception as e:\n return f\"Error: {e}\"\n\ndef run_edit(path: str, old_text: str, new_text: str) -> str:\n try:\n fp = safe_path(path)\n c = fp.read_text()\n if old_text not in c:\n return f\"Error: Text not found in {path}\"\n fp.write_text(c.replace(old_text, new_text, 1))\n return f\"Edited {path}\"\n except Exception as e:\n return f\"Error: {e}\"\n\n\nTOOL_HANDLERS = {\n \"bash\": lambda **kw: run_bash(kw[\"command\"]),\n \"read_file\": lambda **kw: run_read(kw[\"path\"], kw.get(\"limit\")),\n \"write_file\": lambda **kw: run_write(kw[\"path\"], kw[\"content\"]),\n \"edit_file\": lambda **kw: run_edit(kw[\"path\"], kw[\"old_text\"], kw[\"new_text\"]),\n \"background_run\": lambda **kw: BG.run(kw[\"command\"]),\n \"check_background\": lambda **kw: BG.check(kw.get(\"task_id\")),\n}\n\nTOOLS = [\n {\"name\": \"bash\", \"description\": \"Run a shell command (blocking).\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"command\": {\"type\": \"string\"}}, \"required\": [\"command\"]}},\n {\"name\": \"read_file\", \"description\": \"Read file contents.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"limit\": {\"type\": \"integer\"}}, \"required\": [\"path\"]}},\n {\"name\": \"write_file\", \"description\": \"Write content to file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"content\": {\"type\": \"string\"}}, \"required\": [\"path\", \"content\"]}},\n {\"name\": \"edit_file\", \"description\": \"Replace exact text in file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"old_text\": {\"type\": \"string\"}, \"new_text\": {\"type\": \"string\"}}, \"required\": [\"path\", \"old_text\", \"new_text\"]}},\n {\"name\": \"background_run\", \"description\": \"Run command in background thread. Returns task_id immediately.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"command\": {\"type\": \"string\"}}, \"required\": [\"command\"]}},\n {\"name\": \"check_background\", \"description\": \"Check background task status. Omit task_id to list all.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"task_id\": {\"type\": \"string\"}}}},\n]\n\n\ndef agent_loop(messages: list):\n while True:\n # Drain background notifications and inject as a synthetic user/assistant\n # transcript pair before the next model call (teaching demo behavior).\n notifs = BG.drain_notifications()\n if notifs and messages:\n notif_text = \"\\n\".join(\n f\"[bg:{n['task_id']}] {n['status']}: {n['preview']} \"\n f\"(output_file={n['output_file']})\"\n for n in notifs\n )\n messages.append({\"role\": \"user\", \"content\": f\"\\n{notif_text}\\n\"})\n response = client.messages.create(\n model=MODEL, system=SYSTEM, messages=messages,\n tools=TOOLS, max_tokens=8000,\n )\n messages.append({\"role\": \"assistant\", \"content\": response.content})\n if response.stop_reason != \"tool_use\":\n return\n results = []\n for block in response.content:\n if block.type == \"tool_use\":\n handler = TOOL_HANDLERS.get(block.name)\n try:\n output = handler(**block.input) if handler else f\"Unknown tool: {block.name}\"\n except Exception as e:\n output = f\"Error: {e}\"\n print(f\"> {block.name}:\")\n print(str(output)[:200])\n results.append({\"type\": \"tool_result\", \"tool_use_id\": block.id, \"content\": str(output)})\n messages.append({\"role\": \"user\", \"content\": results})\n\n\nif __name__ == \"__main__\":\n history = []\n while True:\n try:\n query = input(\"\\033[36ms13 >> \\033[0m\")\n except (EOFError, KeyboardInterrupt):\n break\n if query.strip().lower() in (\"q\", \"exit\", \"\"):\n break\n history.append({\"role\": \"user\", \"content\": query})\n agent_loop(history)\n response_content = history[-1][\"content\"]\n if isinstance(response_content, list):\n for block in response_content:\n if hasattr(block, \"text\"):\n print(block.text)\n print()\n" }, { "id": "s14", @@ -980,7 +980,7 @@ "filename": "s15_agent_teams.py", "title": "Agent Teams", "subtitle": "Persistent Specialists", - "loc": 353, + "loc": 350, "tools": [ "alice", "bash", @@ -1048,14 +1048,14 @@ } ], "layer": "platform", - "source": "#!/usr/bin/env python3\n# Harness: team mailboxes -- multiple models, coordinated through files.\n\"\"\"\ns15_agent_teams.py - Agent Teams\n\nPersistent named agents with file-based JSONL inboxes. Each teammate runs\nits own agent loop in a separate thread. Communication happens through\nappend-only inbox files.\n\n Subagent (s04): spawn -> execute -> return summary -> destroyed\n Teammate (s15): spawn -> work -> idle -> work -> ... -> shutdown\n\n .team/config.json .team/inbox/\n +----------------------------+ +------------------+\n | {\"team_name\": \"default\", | | alice.jsonl |\n | \"members\": [ | | bob.jsonl |\n | {\"name\":\"alice\", | | lead.jsonl |\n | \"role\":\"coder\", | +------------------+\n | \"status\":\"idle\"} |\n | ]} | send_message(\"alice\", \"fix bug\"):\n +----------------------------+ open(\"alice.jsonl\", \"a\").write(msg)\n\n read_inbox(\"alice\"):\n spawn_teammate(\"alice\",\"coder\",...) msgs = [json.loads(l) for l in ...]\n | open(\"alice.jsonl\", \"w\").close()\n v return msgs # drain\n Thread: alice Thread: bob\n +------------------+ +------------------+\n | agent_loop | | agent_loop |\n | status: working | | status: idle |\n | ... runs tools | | ... waits ... |\n | status -> idle | | |\n +------------------+ +------------------+\n\nKey idea: teammates have names, inboxes, and independent loops.\n\nRead this file in this order:\n1. MessageBus: how messages are queued and drained.\n2. TeammateManager: what persistent teammate state looks like.\n3. _teammate_loop / TOOL_HANDLERS: how each named teammate keeps re-entering the same tool loop.\n\nMost common confusion:\n- a teammate is not a one-shot subagent\n- an inbox message is not yet a full protocol request\n\nTeaching boundary:\nthis file teaches persistent named workers plus mailboxes.\nApproval protocols and autonomous policies are added in later chapters.\n\"\"\"\n\nimport json\nimport os\nimport subprocess\nimport threading\nimport time\nfrom pathlib import Path\n\nfrom anthropic import Anthropic\nfrom dotenv import load_dotenv\n\nload_dotenv(override=True)\nif os.getenv(\"ANTHROPIC_BASE_URL\"):\n os.environ.pop(\"ANTHROPIC_AUTH_TOKEN\", None)\n\nWORKDIR = Path.cwd()\nclient = Anthropic(base_url=os.getenv(\"ANTHROPIC_BASE_URL\"))\nMODEL = os.environ[\"MODEL_ID\"]\nTEAM_DIR = WORKDIR / \".team\"\nINBOX_DIR = TEAM_DIR / \"inbox\"\n\nSYSTEM = f\"You are a team lead at {WORKDIR}. Spawn teammates and communicate via inboxes.\"\n\nVALID_MSG_TYPES = {\n \"message\",\n \"broadcast\",\n \"shutdown_request\",\n \"shutdown_response\",\n \"plan_approval\",\n \"plan_approval_response\",\n}\n\n\n# -- MessageBus: JSONL inbox per teammate --\nclass MessageBus:\n def __init__(self, inbox_dir: Path):\n self.dir = inbox_dir\n self.dir.mkdir(parents=True, exist_ok=True)\n\n def send(self, sender: str, to: str, content: str,\n msg_type: str = \"message\", extra: dict = None) -> str:\n if msg_type not in VALID_MSG_TYPES:\n return f\"Error: Invalid type '{msg_type}'. Valid: {VALID_MSG_TYPES}\"\n msg = {\n \"type\": msg_type,\n \"from\": sender,\n \"content\": content,\n \"timestamp\": time.time(),\n }\n if extra:\n msg.update(extra)\n inbox_path = self.dir / f\"{to}.jsonl\"\n with open(inbox_path, \"a\") as f:\n f.write(json.dumps(msg) + \"\\n\")\n return f\"Sent {msg_type} to {to}\"\n\n def read_inbox(self, name: str) -> list:\n inbox_path = self.dir / f\"{name}.jsonl\"\n if not inbox_path.exists():\n return []\n messages = []\n for line in inbox_path.read_text().strip().splitlines():\n if line:\n messages.append(json.loads(line))\n inbox_path.write_text(\"\")\n return messages\n\n def broadcast(self, sender: str, content: str, teammates: list) -> str:\n count = 0\n for name in teammates:\n if name != sender:\n self.send(sender, name, content, \"broadcast\")\n count += 1\n return f\"Broadcast to {count} teammates\"\n\n\nBUS = MessageBus(INBOX_DIR)\n\n\n# -- TeammateManager: persistent named agents with config.json --\nclass TeammateManager:\n \"\"\"Persistent teammate registry plus worker-loop launcher.\"\"\"\n\n def __init__(self, team_dir: Path):\n self.dir = team_dir\n self.dir.mkdir(exist_ok=True)\n self.config_path = self.dir / \"config.json\"\n self.config = self._load_config()\n self.threads = {}\n\n def _load_config(self) -> dict:\n if self.config_path.exists():\n return json.loads(self.config_path.read_text())\n return {\"team_name\": \"default\", \"members\": []}\n\n def _save_config(self):\n self.config_path.write_text(json.dumps(self.config, indent=2))\n\n def _find_member(self, name: str) -> dict:\n for m in self.config[\"members\"]:\n if m[\"name\"] == name:\n return m\n return None\n\n def spawn(self, name: str, role: str, prompt: str) -> str:\n member = self._find_member(name)\n if member:\n if member[\"status\"] not in (\"idle\", \"shutdown\"):\n return f\"Error: '{name}' is currently {member['status']}\"\n member[\"status\"] = \"working\"\n member[\"role\"] = role\n else:\n member = {\"name\": name, \"role\": role, \"status\": \"working\"}\n self.config[\"members\"].append(member)\n self._save_config()\n thread = threading.Thread(\n target=self._teammate_loop,\n args=(name, role, prompt),\n daemon=True,\n )\n self.threads[name] = thread\n thread.start()\n return f\"Spawned '{name}' (role: {role})\"\n\n def _teammate_loop(self, name: str, role: str, prompt: str):\n sys_prompt = (\n f\"You are '{name}', role: {role}, at {WORKDIR}. \"\n f\"Use send_message to communicate. Complete your task.\"\n )\n messages = [{\"role\": \"user\", \"content\": prompt}]\n tools = self._teammate_tools()\n for _ in range(50):\n inbox = BUS.read_inbox(name)\n for msg in inbox:\n messages.append({\"role\": \"user\", \"content\": json.dumps(msg)})\n try:\n response = client.messages.create(\n model=MODEL,\n system=sys_prompt,\n messages=messages,\n tools=tools,\n max_tokens=8000,\n )\n except Exception:\n break\n messages.append({\"role\": \"assistant\", \"content\": response.content})\n if response.stop_reason != \"tool_use\":\n break\n results = []\n for block in response.content:\n if block.type == \"tool_use\":\n output = self._exec(name, block.name, block.input)\n print(f\" [{name}] {block.name}: {str(output)[:120]}\")\n results.append({\n \"type\": \"tool_result\",\n \"tool_use_id\": block.id,\n \"content\": str(output),\n })\n messages.append({\"role\": \"user\", \"content\": results})\n member = self._find_member(name)\n if member and member[\"status\"] != \"shutdown\":\n member[\"status\"] = \"idle\"\n self._save_config()\n\n def _exec(self, sender: str, tool_name: str, args: dict) -> str:\n # these base tools are unchanged from s02\n if tool_name == \"bash\":\n return _run_bash(args[\"command\"])\n if tool_name == \"read_file\":\n return _run_read(args[\"path\"])\n if tool_name == \"write_file\":\n return _run_write(args[\"path\"], args[\"content\"])\n if tool_name == \"edit_file\":\n return _run_edit(args[\"path\"], args[\"old_text\"], args[\"new_text\"])\n if tool_name == \"send_message\":\n return BUS.send(sender, args[\"to\"], args[\"content\"], args.get(\"msg_type\", \"message\"))\n if tool_name == \"read_inbox\":\n return json.dumps(BUS.read_inbox(sender), indent=2)\n return f\"Unknown tool: {tool_name}\"\n\n def _teammate_tools(self) -> list:\n # these base tools are unchanged from s02\n return [\n {\"name\": \"bash\", \"description\": \"Run a shell command.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"command\": {\"type\": \"string\"}}, \"required\": [\"command\"]}},\n {\"name\": \"read_file\", \"description\": \"Read file contents.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}}, \"required\": [\"path\"]}},\n {\"name\": \"write_file\", \"description\": \"Write content to file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"content\": {\"type\": \"string\"}}, \"required\": [\"path\", \"content\"]}},\n {\"name\": \"edit_file\", \"description\": \"Replace exact text in file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"old_text\": {\"type\": \"string\"}, \"new_text\": {\"type\": \"string\"}}, \"required\": [\"path\", \"old_text\", \"new_text\"]}},\n {\"name\": \"send_message\", \"description\": \"Send message to a teammate.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"to\": {\"type\": \"string\"}, \"content\": {\"type\": \"string\"}, \"msg_type\": {\"type\": \"string\", \"enum\": list(VALID_MSG_TYPES)}}, \"required\": [\"to\", \"content\"]}},\n {\"name\": \"read_inbox\", \"description\": \"Read and drain your inbox.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {}}},\n ]\n\n def list_all(self) -> str:\n if not self.config[\"members\"]:\n return \"No teammates.\"\n lines = [f\"Team: {self.config['team_name']}\"]\n for m in self.config[\"members\"]:\n lines.append(f\" {m['name']} ({m['role']}): {m['status']}\")\n return \"\\n\".join(lines)\n\n def member_names(self) -> list:\n return [m[\"name\"] for m in self.config[\"members\"]]\n\n\nTEAM = TeammateManager(TEAM_DIR)\n\n\n# -- Base tool implementations (these base tools are unchanged from s02) --\ndef _safe_path(p: str) -> Path:\n path = (WORKDIR / p).resolve()\n if not path.is_relative_to(WORKDIR):\n raise ValueError(f\"Path escapes workspace: {p}\")\n return path\n\n\ndef _run_bash(command: str) -> str:\n dangerous = [\"rm -rf /\", \"sudo\", \"shutdown\", \"reboot\"]\n if any(d in command for d in dangerous):\n return \"Error: Dangerous command blocked\"\n try:\n r = subprocess.run(\n command, shell=True, cwd=WORKDIR,\n capture_output=True, text=True, timeout=120,\n )\n out = (r.stdout + r.stderr).strip()\n return out[:50000] if out else \"(no output)\"\n except subprocess.TimeoutExpired:\n return \"Error: Timeout (120s)\"\n\n\ndef _run_read(path: str, limit: int = None) -> str:\n try:\n lines = _safe_path(path).read_text().splitlines()\n if limit and limit < len(lines):\n lines = lines[:limit] + [f\"... ({len(lines) - limit} more)\"]\n return \"\\n\".join(lines)[:50000]\n except Exception as e:\n return f\"Error: {e}\"\n\n\ndef _run_write(path: str, content: str) -> str:\n try:\n fp = _safe_path(path)\n fp.parent.mkdir(parents=True, exist_ok=True)\n fp.write_text(content)\n return f\"Wrote {len(content)} bytes\"\n except Exception as e:\n return f\"Error: {e}\"\n\n\ndef _run_edit(path: str, old_text: str, new_text: str) -> str:\n try:\n fp = _safe_path(path)\n c = fp.read_text()\n if old_text not in c:\n return f\"Error: Text not found in {path}\"\n fp.write_text(c.replace(old_text, new_text, 1))\n return f\"Edited {path}\"\n except Exception as e:\n return f\"Error: {e}\"\n\n\n# -- Lead tool dispatch (9 tools) --\nTOOL_HANDLERS = {\n \"bash\": lambda **kw: _run_bash(kw[\"command\"]),\n \"read_file\": lambda **kw: _run_read(kw[\"path\"], kw.get(\"limit\")),\n \"write_file\": lambda **kw: _run_write(kw[\"path\"], kw[\"content\"]),\n \"edit_file\": lambda **kw: _run_edit(kw[\"path\"], kw[\"old_text\"], kw[\"new_text\"]),\n \"spawn_teammate\": lambda **kw: TEAM.spawn(kw[\"name\"], kw[\"role\"], kw[\"prompt\"]),\n \"list_teammates\": lambda **kw: TEAM.list_all(),\n \"send_message\": lambda **kw: BUS.send(\"lead\", kw[\"to\"], kw[\"content\"], kw.get(\"msg_type\", \"message\")),\n \"read_inbox\": lambda **kw: json.dumps(BUS.read_inbox(\"lead\"), indent=2),\n \"broadcast\": lambda **kw: BUS.broadcast(\"lead\", kw[\"content\"], TEAM.member_names()),\n}\n\n# these base tools are unchanged from s02\nTOOLS = [\n {\"name\": \"bash\", \"description\": \"Run a shell command.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"command\": {\"type\": \"string\"}}, \"required\": [\"command\"]}},\n {\"name\": \"read_file\", \"description\": \"Read file contents.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"limit\": {\"type\": \"integer\"}}, \"required\": [\"path\"]}},\n {\"name\": \"write_file\", \"description\": \"Write content to file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"content\": {\"type\": \"string\"}}, \"required\": [\"path\", \"content\"]}},\n {\"name\": \"edit_file\", \"description\": \"Replace exact text in file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"old_text\": {\"type\": \"string\"}, \"new_text\": {\"type\": \"string\"}}, \"required\": [\"path\", \"old_text\", \"new_text\"]}},\n {\"name\": \"spawn_teammate\", \"description\": \"Spawn a persistent teammate that runs in its own thread.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"name\": {\"type\": \"string\"}, \"role\": {\"type\": \"string\"}, \"prompt\": {\"type\": \"string\"}}, \"required\": [\"name\", \"role\", \"prompt\"]}},\n {\"name\": \"list_teammates\", \"description\": \"List all teammates with name, role, status.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {}}},\n {\"name\": \"send_message\", \"description\": \"Send a message to a teammate's inbox.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"to\": {\"type\": \"string\"}, \"content\": {\"type\": \"string\"}, \"msg_type\": {\"type\": \"string\", \"enum\": list(VALID_MSG_TYPES)}}, \"required\": [\"to\", \"content\"]}},\n {\"name\": \"read_inbox\", \"description\": \"Read and drain the lead's inbox.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {}}},\n {\"name\": \"broadcast\", \"description\": \"Send a message to all teammates.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"content\": {\"type\": \"string\"}}, \"required\": [\"content\"]}},\n]\n\n\ndef agent_loop(messages: list):\n while True:\n inbox = BUS.read_inbox(\"lead\")\n if inbox:\n messages.append({\n \"role\": \"user\",\n \"content\": f\"{json.dumps(inbox, indent=2)}\",\n })\n messages.append({\n \"role\": \"assistant\",\n \"content\": \"Noted inbox messages.\",\n })\n response = client.messages.create(\n model=MODEL,\n system=SYSTEM,\n messages=messages,\n tools=TOOLS,\n max_tokens=8000,\n )\n messages.append({\"role\": \"assistant\", \"content\": response.content})\n if response.stop_reason != \"tool_use\":\n return\n results = []\n for block in response.content:\n if block.type == \"tool_use\":\n handler = TOOL_HANDLERS.get(block.name)\n try:\n output = handler(**block.input) if handler else f\"Unknown tool: {block.name}\"\n except Exception as e:\n output = f\"Error: {e}\"\n print(f\"> {block.name}: {str(output)[:200]}\")\n results.append({\n \"type\": \"tool_result\",\n \"tool_use_id\": block.id,\n \"content\": str(output),\n })\n messages.append({\"role\": \"user\", \"content\": results})\n\n\nif __name__ == \"__main__\":\n history = []\n while True:\n try:\n query = input(\"\\033[36ms15 >> \\033[0m\")\n except (EOFError, KeyboardInterrupt):\n break\n if query.strip().lower() in (\"q\", \"exit\", \"\"):\n break\n if query.strip() == \"/team\":\n print(TEAM.list_all())\n continue\n if query.strip() == \"/inbox\":\n print(json.dumps(BUS.read_inbox(\"lead\"), indent=2))\n continue\n history.append({\"role\": \"user\", \"content\": query})\n agent_loop(history)\n response_content = history[-1][\"content\"]\n if isinstance(response_content, list):\n for block in response_content:\n if hasattr(block, \"text\"):\n print(block.text)\n print()\n" + "source": "#!/usr/bin/env python3\n# Harness: team mailboxes -- multiple models, coordinated through files.\n\"\"\"\ns15_agent_teams.py - Agent Teams\n\nPersistent named agents with file-based JSONL inboxes. Each teammate runs\nits own agent loop in a separate thread. Communication happens through\nappend-only inbox files.\n\n Subagent (s04): spawn -> execute -> return summary -> destroyed\n Teammate (s15): spawn -> work -> idle -> work -> ... -> shutdown\n\n .team/config.json .team/inbox/\n +----------------------------+ +------------------+\n | {\"team_name\": \"default\", | | alice.jsonl |\n | \"members\": [ | | bob.jsonl |\n | {\"name\":\"alice\", | | lead.jsonl |\n | \"role\":\"coder\", | +------------------+\n | \"status\":\"idle\"} |\n | ]} | send_message(\"alice\", \"fix bug\"):\n +----------------------------+ open(\"alice.jsonl\", \"a\").write(msg)\n\n read_inbox(\"alice\"):\n spawn_teammate(\"alice\",\"coder\",...) msgs = [json.loads(l) for l in ...]\n | open(\"alice.jsonl\", \"w\").close()\n v return msgs # drain\n Thread: alice Thread: bob\n +------------------+ +------------------+\n | agent_loop | | agent_loop |\n | status: working | | status: idle |\n | ... runs tools | | ... waits ... |\n | status -> idle | | |\n +------------------+ +------------------+\n\nKey idea: teammates have names, inboxes, and independent loops.\n\nRead this file in this order:\n1. MessageBus: how messages are queued and drained.\n2. TeammateManager: what persistent teammate state looks like.\n3. _teammate_loop / TOOL_HANDLERS: how each named teammate keeps re-entering the same tool loop.\n\nMost common confusion:\n- a teammate is not a one-shot subagent\n- an inbox message is not yet a full protocol request\n\nTeaching boundary:\nthis file teaches persistent named workers plus mailboxes.\nApproval protocols and autonomous policies are added in later chapters.\n\"\"\"\n\nimport json\nimport os\nimport subprocess\nimport threading\nimport time\nfrom pathlib import Path\n\nfrom anthropic import Anthropic\nfrom dotenv import load_dotenv\n\nload_dotenv(override=True)\nif os.getenv(\"ANTHROPIC_BASE_URL\"):\n os.environ.pop(\"ANTHROPIC_AUTH_TOKEN\", None)\n\nWORKDIR = Path.cwd()\nclient = Anthropic(base_url=os.getenv(\"ANTHROPIC_BASE_URL\"))\nMODEL = os.environ[\"MODEL_ID\"]\nTEAM_DIR = WORKDIR / \".team\"\nINBOX_DIR = TEAM_DIR / \"inbox\"\n\nSYSTEM = f\"You are a team lead at {WORKDIR}. Spawn teammates and communicate via inboxes.\"\n\nVALID_MSG_TYPES = {\n \"message\",\n \"broadcast\",\n \"shutdown_request\",\n \"shutdown_response\",\n \"plan_approval\",\n \"plan_approval_response\",\n}\n\n\n# -- MessageBus: JSONL inbox per teammate --\nclass MessageBus:\n def __init__(self, inbox_dir: Path):\n self.dir = inbox_dir\n self.dir.mkdir(parents=True, exist_ok=True)\n\n def send(self, sender: str, to: str, content: str,\n msg_type: str = \"message\", extra: dict = None) -> str:\n if msg_type not in VALID_MSG_TYPES:\n return f\"Error: Invalid type '{msg_type}'. Valid: {VALID_MSG_TYPES}\"\n msg = {\n \"type\": msg_type,\n \"from\": sender,\n \"content\": content,\n \"timestamp\": time.time(),\n }\n if extra:\n msg.update(extra)\n inbox_path = self.dir / f\"{to}.jsonl\"\n with open(inbox_path, \"a\") as f:\n f.write(json.dumps(msg) + \"\\n\")\n return f\"Sent {msg_type} to {to}\"\n\n def read_inbox(self, name: str) -> list:\n inbox_path = self.dir / f\"{name}.jsonl\"\n if not inbox_path.exists():\n return []\n messages = []\n for line in inbox_path.read_text().strip().splitlines():\n if line:\n messages.append(json.loads(line))\n inbox_path.write_text(\"\")\n return messages\n\n def broadcast(self, sender: str, content: str, teammates: list) -> str:\n count = 0\n for name in teammates:\n if name != sender:\n self.send(sender, name, content, \"broadcast\")\n count += 1\n return f\"Broadcast to {count} teammates\"\n\n\nBUS = MessageBus(INBOX_DIR)\n\n\n# -- TeammateManager: persistent named agents with config.json --\nclass TeammateManager:\n \"\"\"Persistent teammate registry plus worker-loop launcher.\"\"\"\n\n def __init__(self, team_dir: Path):\n self.dir = team_dir\n self.dir.mkdir(exist_ok=True)\n self.config_path = self.dir / \"config.json\"\n self.config = self._load_config()\n self.threads = {}\n\n def _load_config(self) -> dict:\n if self.config_path.exists():\n return json.loads(self.config_path.read_text())\n return {\"team_name\": \"default\", \"members\": []}\n\n def _save_config(self):\n self.config_path.write_text(json.dumps(self.config, indent=2))\n\n def _find_member(self, name: str) -> dict:\n for m in self.config[\"members\"]:\n if m[\"name\"] == name:\n return m\n return None\n\n def spawn(self, name: str, role: str, prompt: str) -> str:\n member = self._find_member(name)\n if member:\n if member[\"status\"] not in (\"idle\", \"shutdown\"):\n return f\"Error: '{name}' is currently {member['status']}\"\n member[\"status\"] = \"working\"\n member[\"role\"] = role\n else:\n member = {\"name\": name, \"role\": role, \"status\": \"working\"}\n self.config[\"members\"].append(member)\n self._save_config()\n thread = threading.Thread(\n target=self._teammate_loop,\n args=(name, role, prompt),\n daemon=True,\n )\n self.threads[name] = thread\n thread.start()\n return f\"Spawned '{name}' (role: {role})\"\n\n def _teammate_loop(self, name: str, role: str, prompt: str):\n sys_prompt = (\n f\"You are '{name}', role: {role}, at {WORKDIR}. \"\n f\"Use send_message to communicate. Complete your task.\"\n )\n messages = [{\"role\": \"user\", \"content\": prompt}]\n tools = self._teammate_tools()\n for _ in range(50):\n inbox = BUS.read_inbox(name)\n for msg in inbox:\n messages.append({\"role\": \"user\", \"content\": json.dumps(msg)})\n try:\n response = client.messages.create(\n model=MODEL,\n system=sys_prompt,\n messages=messages,\n tools=tools,\n max_tokens=8000,\n )\n except Exception:\n break\n messages.append({\"role\": \"assistant\", \"content\": response.content})\n if response.stop_reason != \"tool_use\":\n break\n results = []\n for block in response.content:\n if block.type == \"tool_use\":\n output = self._exec(name, block.name, block.input)\n print(f\" [{name}] {block.name}: {str(output)[:120]}\")\n results.append({\n \"type\": \"tool_result\",\n \"tool_use_id\": block.id,\n \"content\": str(output),\n })\n messages.append({\"role\": \"user\", \"content\": results})\n member = self._find_member(name)\n if member and member[\"status\"] != \"shutdown\":\n member[\"status\"] = \"idle\"\n self._save_config()\n\n def _exec(self, sender: str, tool_name: str, args: dict) -> str:\n # these base tools are unchanged from s02\n if tool_name == \"bash\":\n return _run_bash(args[\"command\"])\n if tool_name == \"read_file\":\n return _run_read(args[\"path\"])\n if tool_name == \"write_file\":\n return _run_write(args[\"path\"], args[\"content\"])\n if tool_name == \"edit_file\":\n return _run_edit(args[\"path\"], args[\"old_text\"], args[\"new_text\"])\n if tool_name == \"send_message\":\n return BUS.send(sender, args[\"to\"], args[\"content\"], args.get(\"msg_type\", \"message\"))\n if tool_name == \"read_inbox\":\n return json.dumps(BUS.read_inbox(sender), indent=2)\n return f\"Unknown tool: {tool_name}\"\n\n def _teammate_tools(self) -> list:\n # these base tools are unchanged from s02\n return [\n {\"name\": \"bash\", \"description\": \"Run a shell command.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"command\": {\"type\": \"string\"}}, \"required\": [\"command\"]}},\n {\"name\": \"read_file\", \"description\": \"Read file contents.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}}, \"required\": [\"path\"]}},\n {\"name\": \"write_file\", \"description\": \"Write content to file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"content\": {\"type\": \"string\"}}, \"required\": [\"path\", \"content\"]}},\n {\"name\": \"edit_file\", \"description\": \"Replace exact text in file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"old_text\": {\"type\": \"string\"}, \"new_text\": {\"type\": \"string\"}}, \"required\": [\"path\", \"old_text\", \"new_text\"]}},\n {\"name\": \"send_message\", \"description\": \"Send message to a teammate.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"to\": {\"type\": \"string\"}, \"content\": {\"type\": \"string\"}, \"msg_type\": {\"type\": \"string\", \"enum\": list(VALID_MSG_TYPES)}}, \"required\": [\"to\", \"content\"]}},\n {\"name\": \"read_inbox\", \"description\": \"Read and drain your inbox.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {}}},\n ]\n\n def list_all(self) -> str:\n if not self.config[\"members\"]:\n return \"No teammates.\"\n lines = [f\"Team: {self.config['team_name']}\"]\n for m in self.config[\"members\"]:\n lines.append(f\" {m['name']} ({m['role']}): {m['status']}\")\n return \"\\n\".join(lines)\n\n def member_names(self) -> list:\n return [m[\"name\"] for m in self.config[\"members\"]]\n\n\nTEAM = TeammateManager(TEAM_DIR)\n\n\n# -- Base tool implementations (these base tools are unchanged from s02) --\ndef _safe_path(p: str) -> Path:\n path = (WORKDIR / p).resolve()\n if not path.is_relative_to(WORKDIR):\n raise ValueError(f\"Path escapes workspace: {p}\")\n return path\n\n\ndef _run_bash(command: str) -> str:\n dangerous = [\"rm -rf /\", \"sudo\", \"shutdown\", \"reboot\"]\n if any(d in command for d in dangerous):\n return \"Error: Dangerous command blocked\"\n try:\n r = subprocess.run(\n command, shell=True, cwd=WORKDIR,\n capture_output=True, text=True, timeout=120,\n )\n out = (r.stdout + r.stderr).strip()\n return out[:50000] if out else \"(no output)\"\n except subprocess.TimeoutExpired:\n return \"Error: Timeout (120s)\"\n\n\ndef _run_read(path: str, limit: int = None) -> str:\n try:\n lines = _safe_path(path).read_text().splitlines()\n if limit and limit < len(lines):\n lines = lines[:limit] + [f\"... ({len(lines) - limit} more)\"]\n return \"\\n\".join(lines)[:50000]\n except Exception as e:\n return f\"Error: {e}\"\n\n\ndef _run_write(path: str, content: str) -> str:\n try:\n fp = _safe_path(path)\n fp.parent.mkdir(parents=True, exist_ok=True)\n fp.write_text(content)\n return f\"Wrote {len(content)} bytes\"\n except Exception as e:\n return f\"Error: {e}\"\n\n\ndef _run_edit(path: str, old_text: str, new_text: str) -> str:\n try:\n fp = _safe_path(path)\n c = fp.read_text()\n if old_text not in c:\n return f\"Error: Text not found in {path}\"\n fp.write_text(c.replace(old_text, new_text, 1))\n return f\"Edited {path}\"\n except Exception as e:\n return f\"Error: {e}\"\n\n\n# -- Lead tool dispatch (9 tools) --\nTOOL_HANDLERS = {\n \"bash\": lambda **kw: _run_bash(kw[\"command\"]),\n \"read_file\": lambda **kw: _run_read(kw[\"path\"], kw.get(\"limit\")),\n \"write_file\": lambda **kw: _run_write(kw[\"path\"], kw[\"content\"]),\n \"edit_file\": lambda **kw: _run_edit(kw[\"path\"], kw[\"old_text\"], kw[\"new_text\"]),\n \"spawn_teammate\": lambda **kw: TEAM.spawn(kw[\"name\"], kw[\"role\"], kw[\"prompt\"]),\n \"list_teammates\": lambda **kw: TEAM.list_all(),\n \"send_message\": lambda **kw: BUS.send(\"lead\", kw[\"to\"], kw[\"content\"], kw.get(\"msg_type\", \"message\")),\n \"read_inbox\": lambda **kw: json.dumps(BUS.read_inbox(\"lead\"), indent=2),\n \"broadcast\": lambda **kw: BUS.broadcast(\"lead\", kw[\"content\"], TEAM.member_names()),\n}\n\n# these base tools are unchanged from s02\nTOOLS = [\n {\"name\": \"bash\", \"description\": \"Run a shell command.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"command\": {\"type\": \"string\"}}, \"required\": [\"command\"]}},\n {\"name\": \"read_file\", \"description\": \"Read file contents.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"limit\": {\"type\": \"integer\"}}, \"required\": [\"path\"]}},\n {\"name\": \"write_file\", \"description\": \"Write content to file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"content\": {\"type\": \"string\"}}, \"required\": [\"path\", \"content\"]}},\n {\"name\": \"edit_file\", \"description\": \"Replace exact text in file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"old_text\": {\"type\": \"string\"}, \"new_text\": {\"type\": \"string\"}}, \"required\": [\"path\", \"old_text\", \"new_text\"]}},\n {\"name\": \"spawn_teammate\", \"description\": \"Spawn a persistent teammate that runs in its own thread.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"name\": {\"type\": \"string\"}, \"role\": {\"type\": \"string\"}, \"prompt\": {\"type\": \"string\"}}, \"required\": [\"name\", \"role\", \"prompt\"]}},\n {\"name\": \"list_teammates\", \"description\": \"List all teammates with name, role, status.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {}}},\n {\"name\": \"send_message\", \"description\": \"Send a message to a teammate's inbox.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"to\": {\"type\": \"string\"}, \"content\": {\"type\": \"string\"}, \"msg_type\": {\"type\": \"string\", \"enum\": list(VALID_MSG_TYPES)}}, \"required\": [\"to\", \"content\"]}},\n {\"name\": \"read_inbox\", \"description\": \"Read and drain the lead's inbox.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {}}},\n {\"name\": \"broadcast\", \"description\": \"Send a message to all teammates.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"content\": {\"type\": \"string\"}}, \"required\": [\"content\"]}},\n]\n\n\ndef agent_loop(messages: list):\n while True:\n inbox = BUS.read_inbox(\"lead\")\n if inbox:\n messages.append({\n \"role\": \"user\",\n \"content\": f\"{json.dumps(inbox, indent=2)}\",\n })\n response = client.messages.create(\n model=MODEL,\n system=SYSTEM,\n messages=messages,\n tools=TOOLS,\n max_tokens=8000,\n )\n messages.append({\"role\": \"assistant\", \"content\": response.content})\n if response.stop_reason != \"tool_use\":\n return\n results = []\n for block in response.content:\n if block.type == \"tool_use\":\n handler = TOOL_HANDLERS.get(block.name)\n try:\n output = handler(**block.input) if handler else f\"Unknown tool: {block.name}\"\n except Exception as e:\n output = f\"Error: {e}\"\n print(f\"> {block.name}:\")\n print(str(output)[:200])\n results.append({\n \"type\": \"tool_result\",\n \"tool_use_id\": block.id,\n \"content\": str(output),\n })\n messages.append({\"role\": \"user\", \"content\": results})\n\n\nif __name__ == \"__main__\":\n history = []\n while True:\n try:\n query = input(\"\\033[36ms15 >> \\033[0m\")\n except (EOFError, KeyboardInterrupt):\n break\n if query.strip().lower() in (\"q\", \"exit\", \"\"):\n break\n if query.strip() == \"/team\":\n print(TEAM.list_all())\n continue\n if query.strip() == \"/inbox\":\n print(json.dumps(BUS.read_inbox(\"lead\"), indent=2))\n continue\n history.append({\"role\": \"user\", \"content\": query})\n agent_loop(history)\n response_content = history[-1][\"content\"]\n if isinstance(response_content, list):\n for block in response_content:\n if hasattr(block, \"text\"):\n print(block.text)\n print()\n" }, { "id": "s16", "filename": "s16_team_protocols.py", "title": "Team Protocols", "subtitle": "Shared Request-Response Rules", - "loc": 485, + "loc": 482, "tools": [ "bash", "read_file", @@ -1142,7 +1142,7 @@ } ], "layer": "platform", - "source": "#!/usr/bin/env python3\n# Harness: protocols -- structured handshakes between models.\n\"\"\"\ns16_team_protocols.py - Team Protocols\n\nShutdown protocol and plan approval protocol, both using the same\nrequest_id correlation pattern. Builds on s15's mailbox-based team messaging.\n\n Shutdown FSM: pending -> approved | rejected\n\n Lead Teammate\n +---------------------+ +---------------------+\n | shutdown_request | | |\n | { | -------> | receives request |\n | request_id: abc | | decides: approve? |\n | } | | |\n +---------------------+ +---------------------+\n |\n +---------------------+ +-------v-------------+\n | shutdown_response | <------- | shutdown_response |\n | { | | { |\n | request_id: abc | | request_id: abc |\n | approve: true | | approve: true |\n | } | | } |\n +---------------------+ +---------------------+\n |\n v\n status -> \"shutdown\", thread stops\n\n Plan approval FSM: pending -> approved | rejected\n\n Teammate Lead\n +---------------------+ +---------------------+\n | plan_approval | | |\n | submit: {plan:\"...\"}| -------> | reviews plan text |\n +---------------------+ | approve/reject? |\n +---------------------+\n |\n +---------------------+ +-------v-------------+\n | plan_approval_response| <------ | plan_approval |\n | {approve: true} | | review: {req_id, |\n +---------------------+ | approve: true} |\n +---------------------+\n\n Request store: .team/requests/{request_id}.json\n\nKey idea: one request/response shape can support multiple kinds of team workflow.\nProtocol requests are structured workflow objects, not normal free-form chat.\n\nRead this file in this order:\n1. MessageBus: how protocol envelopes still travel through the same inbox surface.\n2. Request files under .team/requests: how a request keeps durable status after the message is sent.\n3. Protocol handlers: how shutdown and plan approval reuse the same correlation pattern.\n\nMost common confusion:\n- a protocol request is not a normal teammate chat message\n- a request record is not a task record\n\nTeaching boundary:\nthis file teaches durable handshakes first.\nAutonomous claiming, task selection, and worktree assignment stay in later chapters.\n\"\"\"\n\nimport json\nimport os\nimport subprocess\nimport threading\nimport time\nimport uuid\nfrom pathlib import Path\n\nfrom anthropic import Anthropic\nfrom dotenv import load_dotenv\n\nload_dotenv(override=True)\nif os.getenv(\"ANTHROPIC_BASE_URL\"):\n os.environ.pop(\"ANTHROPIC_AUTH_TOKEN\", None)\n\nWORKDIR = Path.cwd()\nclient = Anthropic(base_url=os.getenv(\"ANTHROPIC_BASE_URL\"))\nMODEL = os.environ[\"MODEL_ID\"]\nTEAM_DIR = WORKDIR / \".team\"\nINBOX_DIR = TEAM_DIR / \"inbox\"\nREQUESTS_DIR = TEAM_DIR / \"requests\"\n\nSYSTEM = f\"You are a team lead at {WORKDIR}. Manage teammates with shutdown and plan approval protocols.\"\n\nVALID_MSG_TYPES = {\n \"message\",\n \"broadcast\",\n \"shutdown_request\",\n \"shutdown_response\",\n \"plan_approval\",\n \"plan_approval_response\",\n}\n\n# -- MessageBus: JSONL inbox per teammate --\nclass MessageBus:\n def __init__(self, inbox_dir: Path):\n self.dir = inbox_dir\n self.dir.mkdir(parents=True, exist_ok=True)\n\n def send(self, sender: str, to: str, content: str,\n msg_type: str = \"message\", extra: dict = None) -> str:\n if msg_type not in VALID_MSG_TYPES:\n return f\"Error: Invalid type '{msg_type}'. Valid: {VALID_MSG_TYPES}\"\n msg = {\n \"type\": msg_type,\n \"from\": sender,\n \"content\": content,\n \"timestamp\": time.time(),\n }\n if extra:\n msg.update(extra)\n inbox_path = self.dir / f\"{to}.jsonl\"\n with open(inbox_path, \"a\") as f:\n f.write(json.dumps(msg) + \"\\n\")\n return f\"Sent {msg_type} to {to}\"\n\n def read_inbox(self, name: str) -> list:\n inbox_path = self.dir / f\"{name}.jsonl\"\n if not inbox_path.exists():\n return []\n messages = []\n for line in inbox_path.read_text().strip().splitlines():\n if line:\n messages.append(json.loads(line))\n inbox_path.write_text(\"\")\n return messages\n\n def broadcast(self, sender: str, content: str, teammates: list) -> str:\n count = 0\n for name in teammates:\n if name != sender:\n self.send(sender, name, content, \"broadcast\")\n count += 1\n return f\"Broadcast to {count} teammates\"\n\n\nBUS = MessageBus(INBOX_DIR)\n\n\nclass RequestStore:\n \"\"\"\n Durable request records for protocol workflows.\n\n Protocol state should survive long enough to inspect, resume, or reconcile.\n This store keeps one JSON file per request_id under .team/requests/.\n \"\"\"\n\n def __init__(self, base_dir: Path):\n self.dir = base_dir\n self.dir.mkdir(parents=True, exist_ok=True)\n self._lock = threading.Lock()\n\n def _path(self, request_id: str) -> Path:\n return self.dir / f\"{request_id}.json\"\n\n def create(self, record: dict) -> dict:\n request_id = record[\"request_id\"]\n with self._lock:\n self._path(request_id).write_text(json.dumps(record, indent=2))\n return record\n\n def get(self, request_id: str) -> dict | None:\n path = self._path(request_id)\n if not path.exists():\n return None\n return json.loads(path.read_text())\n\n def update(self, request_id: str, **changes) -> dict | None:\n with self._lock:\n record = self.get(request_id)\n if not record:\n return None\n record.update(changes)\n record[\"updated_at\"] = time.time()\n self._path(request_id).write_text(json.dumps(record, indent=2))\n return record\n\n\nREQUEST_STORE = RequestStore(REQUESTS_DIR)\n\n\n# -- TeammateManager with shutdown + plan approval --\nclass TeammateManager:\n def __init__(self, team_dir: Path):\n self.dir = team_dir\n self.dir.mkdir(exist_ok=True)\n self.config_path = self.dir / \"config.json\"\n self.config = self._load_config()\n self.threads = {}\n\n def _load_config(self) -> dict:\n if self.config_path.exists():\n return json.loads(self.config_path.read_text())\n return {\"team_name\": \"default\", \"members\": []}\n\n def _save_config(self):\n self.config_path.write_text(json.dumps(self.config, indent=2))\n\n def _find_member(self, name: str) -> dict:\n for m in self.config[\"members\"]:\n if m[\"name\"] == name:\n return m\n return None\n\n def spawn(self, name: str, role: str, prompt: str) -> str:\n member = self._find_member(name)\n if member:\n if member[\"status\"] not in (\"idle\", \"shutdown\"):\n return f\"Error: '{name}' is currently {member['status']}\"\n member[\"status\"] = \"working\"\n member[\"role\"] = role\n else:\n member = {\"name\": name, \"role\": role, \"status\": \"working\"}\n self.config[\"members\"].append(member)\n self._save_config()\n thread = threading.Thread(\n target=self._teammate_loop,\n args=(name, role, prompt),\n daemon=True,\n )\n self.threads[name] = thread\n thread.start()\n return f\"Spawned '{name}' (role: {role})\"\n\n def _teammate_loop(self, name: str, role: str, prompt: str):\n sys_prompt = (\n f\"You are '{name}', role: {role}, at {WORKDIR}. \"\n f\"Submit plans via plan_approval before major work. \"\n f\"Respond to shutdown_request with shutdown_response.\"\n )\n messages = [{\"role\": \"user\", \"content\": prompt}]\n tools = self._teammate_tools()\n should_exit = False\n for _ in range(50):\n inbox = BUS.read_inbox(name)\n for msg in inbox:\n messages.append({\"role\": \"user\", \"content\": json.dumps(msg)})\n if should_exit:\n break\n try:\n response = client.messages.create(\n model=MODEL,\n system=sys_prompt,\n messages=messages,\n tools=tools,\n max_tokens=8000,\n )\n except Exception:\n break\n messages.append({\"role\": \"assistant\", \"content\": response.content})\n if response.stop_reason != \"tool_use\":\n break\n results = []\n for block in response.content:\n if block.type == \"tool_use\":\n output = self._exec(name, block.name, block.input)\n print(f\" [{name}] {block.name}: {str(output)[:120]}\")\n results.append({\n \"type\": \"tool_result\",\n \"tool_use_id\": block.id,\n \"content\": str(output),\n })\n if block.name == \"shutdown_response\" and block.input.get(\"approve\"):\n should_exit = True\n messages.append({\"role\": \"user\", \"content\": results})\n member = self._find_member(name)\n if member:\n member[\"status\"] = \"shutdown\" if should_exit else \"idle\"\n self._save_config()\n\n def _exec(self, sender: str, tool_name: str, args: dict) -> str:\n # these base tools are unchanged from s02\n if tool_name == \"bash\":\n return _run_bash(args[\"command\"])\n if tool_name == \"read_file\":\n return _run_read(args[\"path\"])\n if tool_name == \"write_file\":\n return _run_write(args[\"path\"], args[\"content\"])\n if tool_name == \"edit_file\":\n return _run_edit(args[\"path\"], args[\"old_text\"], args[\"new_text\"])\n if tool_name == \"send_message\":\n return BUS.send(sender, args[\"to\"], args[\"content\"], args.get(\"msg_type\", \"message\"))\n if tool_name == \"read_inbox\":\n return json.dumps(BUS.read_inbox(sender), indent=2)\n if tool_name == \"shutdown_response\":\n req_id = args[\"request_id\"]\n approve = args[\"approve\"]\n updated = REQUEST_STORE.update(\n req_id,\n status=\"approved\" if approve else \"rejected\",\n resolved_by=sender,\n resolved_at=time.time(),\n response={\"approve\": approve, \"reason\": args.get(\"reason\", \"\")},\n )\n if not updated:\n return f\"Error: Unknown shutdown request {req_id}\"\n BUS.send(\n sender, \"lead\", args.get(\"reason\", \"\"),\n \"shutdown_response\", {\"request_id\": req_id, \"approve\": approve},\n )\n return f\"Shutdown {'approved' if approve else 'rejected'}\"\n if tool_name == \"plan_approval\":\n plan_text = args.get(\"plan\", \"\")\n req_id = str(uuid.uuid4())[:8]\n REQUEST_STORE.create({\n \"request_id\": req_id,\n \"kind\": \"plan_approval\",\n \"from\": sender,\n \"to\": \"lead\",\n \"status\": \"pending\",\n \"plan\": plan_text,\n \"created_at\": time.time(),\n \"updated_at\": time.time(),\n })\n BUS.send(\n sender, \"lead\", plan_text, \"plan_approval\",\n {\"request_id\": req_id, \"plan\": plan_text},\n )\n return f\"Plan submitted (request_id={req_id}). Waiting for lead approval.\"\n return f\"Unknown tool: {tool_name}\"\n\n def _teammate_tools(self) -> list:\n # these base tools are unchanged from s02\n return [\n {\"name\": \"bash\", \"description\": \"Run a shell command.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"command\": {\"type\": \"string\"}}, \"required\": [\"command\"]}},\n {\"name\": \"read_file\", \"description\": \"Read file contents.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}}, \"required\": [\"path\"]}},\n {\"name\": \"write_file\", \"description\": \"Write content to file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"content\": {\"type\": \"string\"}}, \"required\": [\"path\", \"content\"]}},\n {\"name\": \"edit_file\", \"description\": \"Replace exact text in file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"old_text\": {\"type\": \"string\"}, \"new_text\": {\"type\": \"string\"}}, \"required\": [\"path\", \"old_text\", \"new_text\"]}},\n {\"name\": \"send_message\", \"description\": \"Send message to a teammate.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"to\": {\"type\": \"string\"}, \"content\": {\"type\": \"string\"}, \"msg_type\": {\"type\": \"string\", \"enum\": list(VALID_MSG_TYPES)}}, \"required\": [\"to\", \"content\"]}},\n {\"name\": \"read_inbox\", \"description\": \"Read and drain your inbox.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {}}},\n {\"name\": \"shutdown_response\", \"description\": \"Respond to a shutdown request. Approve to shut down, reject to keep working.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"request_id\": {\"type\": \"string\"}, \"approve\": {\"type\": \"boolean\"}, \"reason\": {\"type\": \"string\"}}, \"required\": [\"request_id\", \"approve\"]}},\n {\"name\": \"plan_approval\", \"description\": \"Submit a plan for lead approval. Provide plan text.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"plan\": {\"type\": \"string\"}}, \"required\": [\"plan\"]}},\n ]\n\n def list_all(self) -> str:\n if not self.config[\"members\"]:\n return \"No teammates.\"\n lines = [f\"Team: {self.config['team_name']}\"]\n for m in self.config[\"members\"]:\n lines.append(f\" {m['name']} ({m['role']}): {m['status']}\")\n return \"\\n\".join(lines)\n\n def member_names(self) -> list:\n return [m[\"name\"] for m in self.config[\"members\"]]\n\n\nTEAM = TeammateManager(TEAM_DIR)\n\n\n# -- Base tool implementations (these base tools are unchanged from s02) --\ndef _safe_path(p: str) -> Path:\n path = (WORKDIR / p).resolve()\n if not path.is_relative_to(WORKDIR):\n raise ValueError(f\"Path escapes workspace: {p}\")\n return path\n\n\ndef _run_bash(command: str) -> str:\n dangerous = [\"rm -rf /\", \"sudo\", \"shutdown\", \"reboot\"]\n if any(d in command for d in dangerous):\n return \"Error: Dangerous command blocked\"\n try:\n r = subprocess.run(\n command, shell=True, cwd=WORKDIR,\n capture_output=True, text=True, timeout=120,\n )\n out = (r.stdout + r.stderr).strip()\n return out[:50000] if out else \"(no output)\"\n except subprocess.TimeoutExpired:\n return \"Error: Timeout (120s)\"\n\n\ndef _run_read(path: str, limit: int = None) -> str:\n try:\n lines = _safe_path(path).read_text().splitlines()\n if limit and limit < len(lines):\n lines = lines[:limit] + [f\"... ({len(lines) - limit} more)\"]\n return \"\\n\".join(lines)[:50000]\n except Exception as e:\n return f\"Error: {e}\"\n\n\ndef _run_write(path: str, content: str) -> str:\n try:\n fp = _safe_path(path)\n fp.parent.mkdir(parents=True, exist_ok=True)\n fp.write_text(content)\n return f\"Wrote {len(content)} bytes\"\n except Exception as e:\n return f\"Error: {e}\"\n\n\ndef _run_edit(path: str, old_text: str, new_text: str) -> str:\n try:\n fp = _safe_path(path)\n c = fp.read_text()\n if old_text not in c:\n return f\"Error: Text not found in {path}\"\n fp.write_text(c.replace(old_text, new_text, 1))\n return f\"Edited {path}\"\n except Exception as e:\n return f\"Error: {e}\"\n\n\n# -- Lead-specific protocol handlers --\ndef handle_shutdown_request(teammate: str) -> str:\n req_id = str(uuid.uuid4())[:8]\n REQUEST_STORE.create({\n \"request_id\": req_id,\n \"kind\": \"shutdown\",\n \"from\": \"lead\",\n \"to\": teammate,\n \"status\": \"pending\",\n \"created_at\": time.time(),\n \"updated_at\": time.time(),\n })\n BUS.send(\n \"lead\", teammate, \"Please shut down gracefully.\",\n \"shutdown_request\", {\"request_id\": req_id},\n )\n return f\"Shutdown request {req_id} sent to '{teammate}' (status: pending)\"\n\n\ndef handle_plan_review(request_id: str, approve: bool, feedback: str = \"\") -> str:\n req = REQUEST_STORE.get(request_id)\n if not req:\n return f\"Error: Unknown plan request_id '{request_id}'\"\n REQUEST_STORE.update(\n request_id,\n status=\"approved\" if approve else \"rejected\",\n reviewed_by=\"lead\",\n resolved_at=time.time(),\n feedback=feedback,\n )\n BUS.send(\n \"lead\", req[\"from\"], feedback, \"plan_approval_response\",\n {\"request_id\": request_id, \"approve\": approve, \"feedback\": feedback},\n )\n return f\"Plan {'approved' if approve else 'rejected'} for '{req['from']}'\"\n\n\ndef _check_shutdown_status(request_id: str) -> str:\n return json.dumps(REQUEST_STORE.get(request_id) or {\"error\": \"not found\"})\n\n\n# -- Lead tool dispatch (12 tools) --\nTOOL_HANDLERS = {\n \"bash\": lambda **kw: _run_bash(kw[\"command\"]),\n \"read_file\": lambda **kw: _run_read(kw[\"path\"], kw.get(\"limit\")),\n \"write_file\": lambda **kw: _run_write(kw[\"path\"], kw[\"content\"]),\n \"edit_file\": lambda **kw: _run_edit(kw[\"path\"], kw[\"old_text\"], kw[\"new_text\"]),\n \"spawn_teammate\": lambda **kw: TEAM.spawn(kw[\"name\"], kw[\"role\"], kw[\"prompt\"]),\n \"list_teammates\": lambda **kw: TEAM.list_all(),\n \"send_message\": lambda **kw: BUS.send(\"lead\", kw[\"to\"], kw[\"content\"], kw.get(\"msg_type\", \"message\")),\n \"read_inbox\": lambda **kw: json.dumps(BUS.read_inbox(\"lead\"), indent=2),\n \"broadcast\": lambda **kw: BUS.broadcast(\"lead\", kw[\"content\"], TEAM.member_names()),\n \"shutdown_request\": lambda **kw: handle_shutdown_request(kw[\"teammate\"]),\n \"shutdown_response\": lambda **kw: _check_shutdown_status(kw.get(\"request_id\", \"\")),\n \"plan_approval\": lambda **kw: handle_plan_review(kw[\"request_id\"], kw[\"approve\"], kw.get(\"feedback\", \"\")),\n}\n\n# these base tools are unchanged from s02\nTOOLS = [\n {\"name\": \"bash\", \"description\": \"Run a shell command.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"command\": {\"type\": \"string\"}}, \"required\": [\"command\"]}},\n {\"name\": \"read_file\", \"description\": \"Read file contents.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"limit\": {\"type\": \"integer\"}}, \"required\": [\"path\"]}},\n {\"name\": \"write_file\", \"description\": \"Write content to file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"content\": {\"type\": \"string\"}}, \"required\": [\"path\", \"content\"]}},\n {\"name\": \"edit_file\", \"description\": \"Replace exact text in file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"old_text\": {\"type\": \"string\"}, \"new_text\": {\"type\": \"string\"}}, \"required\": [\"path\", \"old_text\", \"new_text\"]}},\n {\"name\": \"spawn_teammate\", \"description\": \"Spawn a persistent teammate.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"name\": {\"type\": \"string\"}, \"role\": {\"type\": \"string\"}, \"prompt\": {\"type\": \"string\"}}, \"required\": [\"name\", \"role\", \"prompt\"]}},\n {\"name\": \"list_teammates\", \"description\": \"List all teammates.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {}}},\n {\"name\": \"send_message\", \"description\": \"Send a message to a teammate.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"to\": {\"type\": \"string\"}, \"content\": {\"type\": \"string\"}, \"msg_type\": {\"type\": \"string\", \"enum\": list(VALID_MSG_TYPES)}}, \"required\": [\"to\", \"content\"]}},\n {\"name\": \"read_inbox\", \"description\": \"Read and drain the lead's inbox.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {}}},\n {\"name\": \"broadcast\", \"description\": \"Send a message to all teammates.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"content\": {\"type\": \"string\"}}, \"required\": [\"content\"]}},\n {\"name\": \"shutdown_request\", \"description\": \"Request a teammate to shut down gracefully. Returns a request_id for tracking.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"teammate\": {\"type\": \"string\"}}, \"required\": [\"teammate\"]}},\n {\"name\": \"shutdown_response\", \"description\": \"Check the status of a shutdown request by request_id.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"request_id\": {\"type\": \"string\"}}, \"required\": [\"request_id\"]}},\n {\"name\": \"plan_approval\", \"description\": \"Approve or reject a teammate's plan. Provide request_id + approve + optional feedback.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"request_id\": {\"type\": \"string\"}, \"approve\": {\"type\": \"boolean\"}, \"feedback\": {\"type\": \"string\"}}, \"required\": [\"request_id\", \"approve\"]}},\n]\n\n\ndef agent_loop(messages: list):\n while True:\n inbox = BUS.read_inbox(\"lead\")\n if inbox:\n messages.append({\n \"role\": \"user\",\n \"content\": f\"{json.dumps(inbox, indent=2)}\",\n })\n messages.append({\n \"role\": \"assistant\",\n \"content\": \"Noted inbox messages.\",\n })\n response = client.messages.create(\n model=MODEL,\n system=SYSTEM,\n messages=messages,\n tools=TOOLS,\n max_tokens=8000,\n )\n messages.append({\"role\": \"assistant\", \"content\": response.content})\n if response.stop_reason != \"tool_use\":\n return\n results = []\n for block in response.content:\n if block.type == \"tool_use\":\n handler = TOOL_HANDLERS.get(block.name)\n try:\n output = handler(**block.input) if handler else f\"Unknown tool: {block.name}\"\n except Exception as e:\n output = f\"Error: {e}\"\n print(f\"> {block.name}: {str(output)[:200]}\")\n results.append({\n \"type\": \"tool_result\",\n \"tool_use_id\": block.id,\n \"content\": str(output),\n })\n messages.append({\"role\": \"user\", \"content\": results})\n\n\nif __name__ == \"__main__\":\n history = []\n while True:\n try:\n query = input(\"\\033[36ms16 >> \\033[0m\")\n except (EOFError, KeyboardInterrupt):\n break\n if query.strip().lower() in (\"q\", \"exit\", \"\"):\n break\n if query.strip() == \"/team\":\n print(TEAM.list_all())\n continue\n if query.strip() == \"/inbox\":\n print(json.dumps(BUS.read_inbox(\"lead\"), indent=2))\n continue\n history.append({\"role\": \"user\", \"content\": query})\n agent_loop(history)\n response_content = history[-1][\"content\"]\n if isinstance(response_content, list):\n for block in response_content:\n if hasattr(block, \"text\"):\n print(block.text)\n print()\n" + "source": "#!/usr/bin/env python3\n# Harness: protocols -- structured handshakes between models.\n\"\"\"\ns16_team_protocols.py - Team Protocols\n\nShutdown protocol and plan approval protocol, both using the same\nrequest_id correlation pattern. Builds on s15's mailbox-based team messaging.\n\n Shutdown FSM: pending -> approved | rejected\n\n Lead Teammate\n +---------------------+ +---------------------+\n | shutdown_request | | |\n | { | -------> | receives request |\n | request_id: abc | | decides: approve? |\n | } | | |\n +---------------------+ +---------------------+\n |\n +---------------------+ +-------v-------------+\n | shutdown_response | <------- | shutdown_response |\n | { | | { |\n | request_id: abc | | request_id: abc |\n | approve: true | | approve: true |\n | } | | } |\n +---------------------+ +---------------------+\n |\n v\n status -> \"shutdown\", thread stops\n\n Plan approval FSM: pending -> approved | rejected\n\n Teammate Lead\n +---------------------+ +---------------------+\n | plan_approval | | |\n | submit: {plan:\"...\"}| -------> | reviews plan text |\n +---------------------+ | approve/reject? |\n +---------------------+\n |\n +---------------------+ +-------v-------------+\n | plan_approval_response| <------ | plan_approval |\n | {approve: true} | | review: {req_id, |\n +---------------------+ | approve: true} |\n +---------------------+\n\n Request store: .team/requests/{request_id}.json\n\nKey idea: one request/response shape can support multiple kinds of team workflow.\nProtocol requests are structured workflow objects, not normal free-form chat.\n\nRead this file in this order:\n1. MessageBus: how protocol envelopes still travel through the same inbox surface.\n2. Request files under .team/requests: how a request keeps durable status after the message is sent.\n3. Protocol handlers: how shutdown and plan approval reuse the same correlation pattern.\n\nMost common confusion:\n- a protocol request is not a normal teammate chat message\n- a request record is not a task record\n\nTeaching boundary:\nthis file teaches durable handshakes first.\nAutonomous claiming, task selection, and worktree assignment stay in later chapters.\n\"\"\"\n\nimport json\nimport os\nimport subprocess\nimport threading\nimport time\nimport uuid\nfrom pathlib import Path\n\nfrom anthropic import Anthropic\nfrom dotenv import load_dotenv\n\nload_dotenv(override=True)\nif os.getenv(\"ANTHROPIC_BASE_URL\"):\n os.environ.pop(\"ANTHROPIC_AUTH_TOKEN\", None)\n\nWORKDIR = Path.cwd()\nclient = Anthropic(base_url=os.getenv(\"ANTHROPIC_BASE_URL\"))\nMODEL = os.environ[\"MODEL_ID\"]\nTEAM_DIR = WORKDIR / \".team\"\nINBOX_DIR = TEAM_DIR / \"inbox\"\nREQUESTS_DIR = TEAM_DIR / \"requests\"\n\nSYSTEM = f\"You are a team lead at {WORKDIR}. Manage teammates with shutdown and plan approval protocols.\"\n\nVALID_MSG_TYPES = {\n \"message\",\n \"broadcast\",\n \"shutdown_request\",\n \"shutdown_response\",\n \"plan_approval\",\n \"plan_approval_response\",\n}\n\n# -- MessageBus: JSONL inbox per teammate --\nclass MessageBus:\n def __init__(self, inbox_dir: Path):\n self.dir = inbox_dir\n self.dir.mkdir(parents=True, exist_ok=True)\n\n def send(self, sender: str, to: str, content: str,\n msg_type: str = \"message\", extra: dict = None) -> str:\n if msg_type not in VALID_MSG_TYPES:\n return f\"Error: Invalid type '{msg_type}'. Valid: {VALID_MSG_TYPES}\"\n msg = {\n \"type\": msg_type,\n \"from\": sender,\n \"content\": content,\n \"timestamp\": time.time(),\n }\n if extra:\n msg.update(extra)\n inbox_path = self.dir / f\"{to}.jsonl\"\n with open(inbox_path, \"a\") as f:\n f.write(json.dumps(msg) + \"\\n\")\n return f\"Sent {msg_type} to {to}\"\n\n def read_inbox(self, name: str) -> list:\n inbox_path = self.dir / f\"{name}.jsonl\"\n if not inbox_path.exists():\n return []\n messages = []\n for line in inbox_path.read_text().strip().splitlines():\n if line:\n messages.append(json.loads(line))\n inbox_path.write_text(\"\")\n return messages\n\n def broadcast(self, sender: str, content: str, teammates: list) -> str:\n count = 0\n for name in teammates:\n if name != sender:\n self.send(sender, name, content, \"broadcast\")\n count += 1\n return f\"Broadcast to {count} teammates\"\n\n\nBUS = MessageBus(INBOX_DIR)\n\n\nclass RequestStore:\n \"\"\"\n Durable request records for protocol workflows.\n\n Protocol state should survive long enough to inspect, resume, or reconcile.\n This store keeps one JSON file per request_id under .team/requests/.\n \"\"\"\n\n def __init__(self, base_dir: Path):\n self.dir = base_dir\n self.dir.mkdir(parents=True, exist_ok=True)\n self._lock = threading.Lock()\n\n def _path(self, request_id: str) -> Path:\n return self.dir / f\"{request_id}.json\"\n\n def create(self, record: dict) -> dict:\n request_id = record[\"request_id\"]\n with self._lock:\n self._path(request_id).write_text(json.dumps(record, indent=2))\n return record\n\n def get(self, request_id: str) -> dict | None:\n path = self._path(request_id)\n if not path.exists():\n return None\n return json.loads(path.read_text())\n\n def update(self, request_id: str, **changes) -> dict | None:\n with self._lock:\n record = self.get(request_id)\n if not record:\n return None\n record.update(changes)\n record[\"updated_at\"] = time.time()\n self._path(request_id).write_text(json.dumps(record, indent=2))\n return record\n\n\nREQUEST_STORE = RequestStore(REQUESTS_DIR)\n\n\n# -- TeammateManager with shutdown + plan approval --\nclass TeammateManager:\n def __init__(self, team_dir: Path):\n self.dir = team_dir\n self.dir.mkdir(exist_ok=True)\n self.config_path = self.dir / \"config.json\"\n self.config = self._load_config()\n self.threads = {}\n\n def _load_config(self) -> dict:\n if self.config_path.exists():\n return json.loads(self.config_path.read_text())\n return {\"team_name\": \"default\", \"members\": []}\n\n def _save_config(self):\n self.config_path.write_text(json.dumps(self.config, indent=2))\n\n def _find_member(self, name: str) -> dict:\n for m in self.config[\"members\"]:\n if m[\"name\"] == name:\n return m\n return None\n\n def spawn(self, name: str, role: str, prompt: str) -> str:\n member = self._find_member(name)\n if member:\n if member[\"status\"] not in (\"idle\", \"shutdown\"):\n return f\"Error: '{name}' is currently {member['status']}\"\n member[\"status\"] = \"working\"\n member[\"role\"] = role\n else:\n member = {\"name\": name, \"role\": role, \"status\": \"working\"}\n self.config[\"members\"].append(member)\n self._save_config()\n thread = threading.Thread(\n target=self._teammate_loop,\n args=(name, role, prompt),\n daemon=True,\n )\n self.threads[name] = thread\n thread.start()\n return f\"Spawned '{name}' (role: {role})\"\n\n def _teammate_loop(self, name: str, role: str, prompt: str):\n sys_prompt = (\n f\"You are '{name}', role: {role}, at {WORKDIR}. \"\n f\"Submit plans via plan_approval before major work. \"\n f\"Respond to shutdown_request with shutdown_response.\"\n )\n messages = [{\"role\": \"user\", \"content\": prompt}]\n tools = self._teammate_tools()\n should_exit = False\n for _ in range(50):\n inbox = BUS.read_inbox(name)\n for msg in inbox:\n messages.append({\"role\": \"user\", \"content\": json.dumps(msg)})\n if should_exit:\n break\n try:\n response = client.messages.create(\n model=MODEL,\n system=sys_prompt,\n messages=messages,\n tools=tools,\n max_tokens=8000,\n )\n except Exception:\n break\n messages.append({\"role\": \"assistant\", \"content\": response.content})\n if response.stop_reason != \"tool_use\":\n break\n results = []\n for block in response.content:\n if block.type == \"tool_use\":\n output = self._exec(name, block.name, block.input)\n print(f\" [{name}] {block.name}: {str(output)[:120]}\")\n results.append({\n \"type\": \"tool_result\",\n \"tool_use_id\": block.id,\n \"content\": str(output),\n })\n if block.name == \"shutdown_response\" and block.input.get(\"approve\"):\n should_exit = True\n messages.append({\"role\": \"user\", \"content\": results})\n member = self._find_member(name)\n if member:\n member[\"status\"] = \"shutdown\" if should_exit else \"idle\"\n self._save_config()\n\n def _exec(self, sender: str, tool_name: str, args: dict) -> str:\n # these base tools are unchanged from s02\n if tool_name == \"bash\":\n return _run_bash(args[\"command\"])\n if tool_name == \"read_file\":\n return _run_read(args[\"path\"])\n if tool_name == \"write_file\":\n return _run_write(args[\"path\"], args[\"content\"])\n if tool_name == \"edit_file\":\n return _run_edit(args[\"path\"], args[\"old_text\"], args[\"new_text\"])\n if tool_name == \"send_message\":\n return BUS.send(sender, args[\"to\"], args[\"content\"], args.get(\"msg_type\", \"message\"))\n if tool_name == \"read_inbox\":\n return json.dumps(BUS.read_inbox(sender), indent=2)\n if tool_name == \"shutdown_response\":\n req_id = args[\"request_id\"]\n approve = args[\"approve\"]\n updated = REQUEST_STORE.update(\n req_id,\n status=\"approved\" if approve else \"rejected\",\n resolved_by=sender,\n resolved_at=time.time(),\n response={\"approve\": approve, \"reason\": args.get(\"reason\", \"\")},\n )\n if not updated:\n return f\"Error: Unknown shutdown request {req_id}\"\n BUS.send(\n sender, \"lead\", args.get(\"reason\", \"\"),\n \"shutdown_response\", {\"request_id\": req_id, \"approve\": approve},\n )\n return f\"Shutdown {'approved' if approve else 'rejected'}\"\n if tool_name == \"plan_approval\":\n plan_text = args.get(\"plan\", \"\")\n req_id = str(uuid.uuid4())[:8]\n REQUEST_STORE.create({\n \"request_id\": req_id,\n \"kind\": \"plan_approval\",\n \"from\": sender,\n \"to\": \"lead\",\n \"status\": \"pending\",\n \"plan\": plan_text,\n \"created_at\": time.time(),\n \"updated_at\": time.time(),\n })\n BUS.send(\n sender, \"lead\", plan_text, \"plan_approval\",\n {\"request_id\": req_id, \"plan\": plan_text},\n )\n return f\"Plan submitted (request_id={req_id}). Waiting for lead approval.\"\n return f\"Unknown tool: {tool_name}\"\n\n def _teammate_tools(self) -> list:\n # these base tools are unchanged from s02\n return [\n {\"name\": \"bash\", \"description\": \"Run a shell command.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"command\": {\"type\": \"string\"}}, \"required\": [\"command\"]}},\n {\"name\": \"read_file\", \"description\": \"Read file contents.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}}, \"required\": [\"path\"]}},\n {\"name\": \"write_file\", \"description\": \"Write content to file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"content\": {\"type\": \"string\"}}, \"required\": [\"path\", \"content\"]}},\n {\"name\": \"edit_file\", \"description\": \"Replace exact text in file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"old_text\": {\"type\": \"string\"}, \"new_text\": {\"type\": \"string\"}}, \"required\": [\"path\", \"old_text\", \"new_text\"]}},\n {\"name\": \"send_message\", \"description\": \"Send message to a teammate.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"to\": {\"type\": \"string\"}, \"content\": {\"type\": \"string\"}, \"msg_type\": {\"type\": \"string\", \"enum\": list(VALID_MSG_TYPES)}}, \"required\": [\"to\", \"content\"]}},\n {\"name\": \"read_inbox\", \"description\": \"Read and drain your inbox.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {}}},\n {\"name\": \"shutdown_response\", \"description\": \"Respond to a shutdown request. Approve to shut down, reject to keep working.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"request_id\": {\"type\": \"string\"}, \"approve\": {\"type\": \"boolean\"}, \"reason\": {\"type\": \"string\"}}, \"required\": [\"request_id\", \"approve\"]}},\n {\"name\": \"plan_approval\", \"description\": \"Submit a plan for lead approval. Provide plan text.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"plan\": {\"type\": \"string\"}}, \"required\": [\"plan\"]}},\n ]\n\n def list_all(self) -> str:\n if not self.config[\"members\"]:\n return \"No teammates.\"\n lines = [f\"Team: {self.config['team_name']}\"]\n for m in self.config[\"members\"]:\n lines.append(f\" {m['name']} ({m['role']}): {m['status']}\")\n return \"\\n\".join(lines)\n\n def member_names(self) -> list:\n return [m[\"name\"] for m in self.config[\"members\"]]\n\n\nTEAM = TeammateManager(TEAM_DIR)\n\n\n# -- Base tool implementations (these base tools are unchanged from s02) --\ndef _safe_path(p: str) -> Path:\n path = (WORKDIR / p).resolve()\n if not path.is_relative_to(WORKDIR):\n raise ValueError(f\"Path escapes workspace: {p}\")\n return path\n\n\ndef _run_bash(command: str) -> str:\n dangerous = [\"rm -rf /\", \"sudo\", \"shutdown\", \"reboot\"]\n if any(d in command for d in dangerous):\n return \"Error: Dangerous command blocked\"\n try:\n r = subprocess.run(\n command, shell=True, cwd=WORKDIR,\n capture_output=True, text=True, timeout=120,\n )\n out = (r.stdout + r.stderr).strip()\n return out[:50000] if out else \"(no output)\"\n except subprocess.TimeoutExpired:\n return \"Error: Timeout (120s)\"\n\n\ndef _run_read(path: str, limit: int = None) -> str:\n try:\n lines = _safe_path(path).read_text().splitlines()\n if limit and limit < len(lines):\n lines = lines[:limit] + [f\"... ({len(lines) - limit} more)\"]\n return \"\\n\".join(lines)[:50000]\n except Exception as e:\n return f\"Error: {e}\"\n\n\ndef _run_write(path: str, content: str) -> str:\n try:\n fp = _safe_path(path)\n fp.parent.mkdir(parents=True, exist_ok=True)\n fp.write_text(content)\n return f\"Wrote {len(content)} bytes\"\n except Exception as e:\n return f\"Error: {e}\"\n\n\ndef _run_edit(path: str, old_text: str, new_text: str) -> str:\n try:\n fp = _safe_path(path)\n c = fp.read_text()\n if old_text not in c:\n return f\"Error: Text not found in {path}\"\n fp.write_text(c.replace(old_text, new_text, 1))\n return f\"Edited {path}\"\n except Exception as e:\n return f\"Error: {e}\"\n\n\n# -- Lead-specific protocol handlers --\ndef handle_shutdown_request(teammate: str) -> str:\n req_id = str(uuid.uuid4())[:8]\n REQUEST_STORE.create({\n \"request_id\": req_id,\n \"kind\": \"shutdown\",\n \"from\": \"lead\",\n \"to\": teammate,\n \"status\": \"pending\",\n \"created_at\": time.time(),\n \"updated_at\": time.time(),\n })\n BUS.send(\n \"lead\", teammate, \"Please shut down gracefully.\",\n \"shutdown_request\", {\"request_id\": req_id},\n )\n return f\"Shutdown request {req_id} sent to '{teammate}' (status: pending)\"\n\n\ndef handle_plan_review(request_id: str, approve: bool, feedback: str = \"\") -> str:\n req = REQUEST_STORE.get(request_id)\n if not req:\n return f\"Error: Unknown plan request_id '{request_id}'\"\n REQUEST_STORE.update(\n request_id,\n status=\"approved\" if approve else \"rejected\",\n reviewed_by=\"lead\",\n resolved_at=time.time(),\n feedback=feedback,\n )\n BUS.send(\n \"lead\", req[\"from\"], feedback, \"plan_approval_response\",\n {\"request_id\": request_id, \"approve\": approve, \"feedback\": feedback},\n )\n return f\"Plan {'approved' if approve else 'rejected'} for '{req['from']}'\"\n\n\ndef _check_shutdown_status(request_id: str) -> str:\n return json.dumps(REQUEST_STORE.get(request_id) or {\"error\": \"not found\"})\n\n\n# -- Lead tool dispatch (12 tools) --\nTOOL_HANDLERS = {\n \"bash\": lambda **kw: _run_bash(kw[\"command\"]),\n \"read_file\": lambda **kw: _run_read(kw[\"path\"], kw.get(\"limit\")),\n \"write_file\": lambda **kw: _run_write(kw[\"path\"], kw[\"content\"]),\n \"edit_file\": lambda **kw: _run_edit(kw[\"path\"], kw[\"old_text\"], kw[\"new_text\"]),\n \"spawn_teammate\": lambda **kw: TEAM.spawn(kw[\"name\"], kw[\"role\"], kw[\"prompt\"]),\n \"list_teammates\": lambda **kw: TEAM.list_all(),\n \"send_message\": lambda **kw: BUS.send(\"lead\", kw[\"to\"], kw[\"content\"], kw.get(\"msg_type\", \"message\")),\n \"read_inbox\": lambda **kw: json.dumps(BUS.read_inbox(\"lead\"), indent=2),\n \"broadcast\": lambda **kw: BUS.broadcast(\"lead\", kw[\"content\"], TEAM.member_names()),\n \"shutdown_request\": lambda **kw: handle_shutdown_request(kw[\"teammate\"]),\n \"shutdown_response\": lambda **kw: _check_shutdown_status(kw.get(\"request_id\", \"\")),\n \"plan_approval\": lambda **kw: handle_plan_review(kw[\"request_id\"], kw[\"approve\"], kw.get(\"feedback\", \"\")),\n}\n\n# these base tools are unchanged from s02\nTOOLS = [\n {\"name\": \"bash\", \"description\": \"Run a shell command.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"command\": {\"type\": \"string\"}}, \"required\": [\"command\"]}},\n {\"name\": \"read_file\", \"description\": \"Read file contents.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"limit\": {\"type\": \"integer\"}}, \"required\": [\"path\"]}},\n {\"name\": \"write_file\", \"description\": \"Write content to file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"content\": {\"type\": \"string\"}}, \"required\": [\"path\", \"content\"]}},\n {\"name\": \"edit_file\", \"description\": \"Replace exact text in file.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"path\": {\"type\": \"string\"}, \"old_text\": {\"type\": \"string\"}, \"new_text\": {\"type\": \"string\"}}, \"required\": [\"path\", \"old_text\", \"new_text\"]}},\n {\"name\": \"spawn_teammate\", \"description\": \"Spawn a persistent teammate.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"name\": {\"type\": \"string\"}, \"role\": {\"type\": \"string\"}, \"prompt\": {\"type\": \"string\"}}, \"required\": [\"name\", \"role\", \"prompt\"]}},\n {\"name\": \"list_teammates\", \"description\": \"List all teammates.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {}}},\n {\"name\": \"send_message\", \"description\": \"Send a message to a teammate.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"to\": {\"type\": \"string\"}, \"content\": {\"type\": \"string\"}, \"msg_type\": {\"type\": \"string\", \"enum\": list(VALID_MSG_TYPES)}}, \"required\": [\"to\", \"content\"]}},\n {\"name\": \"read_inbox\", \"description\": \"Read and drain the lead's inbox.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {}}},\n {\"name\": \"broadcast\", \"description\": \"Send a message to all teammates.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"content\": {\"type\": \"string\"}}, \"required\": [\"content\"]}},\n {\"name\": \"shutdown_request\", \"description\": \"Request a teammate to shut down gracefully. Returns a request_id for tracking.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"teammate\": {\"type\": \"string\"}}, \"required\": [\"teammate\"]}},\n {\"name\": \"shutdown_response\", \"description\": \"Check the status of a shutdown request by request_id.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"request_id\": {\"type\": \"string\"}}, \"required\": [\"request_id\"]}},\n {\"name\": \"plan_approval\", \"description\": \"Approve or reject a teammate's plan. Provide request_id + approve + optional feedback.\",\n \"input_schema\": {\"type\": \"object\", \"properties\": {\"request_id\": {\"type\": \"string\"}, \"approve\": {\"type\": \"boolean\"}, \"feedback\": {\"type\": \"string\"}}, \"required\": [\"request_id\", \"approve\"]}},\n]\n\n\ndef agent_loop(messages: list):\n while True:\n inbox = BUS.read_inbox(\"lead\")\n if inbox:\n messages.append({\n \"role\": \"user\",\n \"content\": f\"{json.dumps(inbox, indent=2)}\",\n })\n response = client.messages.create(\n model=MODEL,\n system=SYSTEM,\n messages=messages,\n tools=TOOLS,\n max_tokens=8000,\n )\n messages.append({\"role\": \"assistant\", \"content\": response.content})\n if response.stop_reason != \"tool_use\":\n return\n results = []\n for block in response.content:\n if block.type == \"tool_use\":\n handler = TOOL_HANDLERS.get(block.name)\n try:\n output = handler(**block.input) if handler else f\"Unknown tool: {block.name}\"\n except Exception as e:\n output = f\"Error: {e}\"\n print(f\"> {block.name}:\")\n print(str(output)[:200])\n results.append({\n \"type\": \"tool_result\",\n \"tool_use_id\": block.id,\n \"content\": str(output),\n })\n messages.append({\"role\": \"user\", \"content\": results})\n\n\nif __name__ == \"__main__\":\n history = []\n while True:\n try:\n query = input(\"\\033[36ms16 >> \\033[0m\")\n except (EOFError, KeyboardInterrupt):\n break\n if query.strip().lower() in (\"q\", \"exit\", \"\"):\n break\n if query.strip() == \"/team\":\n print(TEAM.list_all())\n continue\n if query.strip() == \"/inbox\":\n print(json.dumps(BUS.read_inbox(\"lead\"), indent=2))\n continue\n history.append({\"role\": \"user\", \"content\": query})\n agent_loop(history)\n response_content = history[-1][\"content\"]\n if isinstance(response_content, list):\n for block in response_content:\n if hasattr(block, \"text\"):\n print(block.text)\n print()\n" }, { "id": "s17", @@ -1475,7 +1475,7 @@ "write_file", "edit_file" ], - "locDelta": 49 + "locDelta": 39 }, { "from": "s02", @@ -1491,7 +1491,7 @@ "newTools": [ "todo" ], - "locDelta": 111 + "locDelta": 110 }, { "from": "s03", @@ -1505,7 +1505,7 @@ "newTools": [ "task" ], - "locDelta": -82 + "locDelta": -79 }, { "from": "s04", @@ -1521,7 +1521,7 @@ "newTools": [ "load_skill" ], - "locDelta": 47 + "locDelta": 44 }, { "from": "s05", @@ -1677,7 +1677,7 @@ "list_teammates", "broadcast" ], - "locDelta": -99 + "locDelta": -102 }, { "from": "s15", @@ -1713,7 +1713,7 @@ "idle", "claim_task" ], - "locDelta": 118 + "locDelta": 121 }, { "from": "s17",