feat: migrate orchestration to n8n and validate lan mvp

2026-03-18 10:05:00 +08:00
parent d2074c3518
commit b145363111
16 changed files with 2429 additions and 254 deletions
--- a/.env.example
+++ b/.env.example
@@ -2,15 +2,25 @@ DEFAULT_EXTERNAL_BASE_URL=http://test.hyzq.net:8081
 LOCAL_OPENAI_BASE_URL=http://127.0.0.1:8317/v1
 LOCAL_OPENAI_MODEL=GLM-5
 LOCAL_OPENAI_API_KEY=
-FASTGPT_BASE_URL=http://127.0.0.1:3000
-FASTGPT_DATASET_API_KEY=
+N8N_BASE_URL=http://127.0.0.1:5670
+N8N_ANALYSIS_WEBHOOK_PATH=/webhook/storyforge-analysis
+N8N_REAL_CUT_WEBHOOK_PATH=/webhook/storyforge-real-cut
+N8N_AI_VIDEO_WEBHOOK_PATH=/webhook/storyforge-ai-video
+ORCHESTRATOR_SHARED_SECRET=storyforge-local-secret
+CUTVIDEO_BASE_URL=
+CUTVIDEO_API_KEY=
+CUTVIDEO_BASE_CONFIG=example.job.yaml
+CUTVIDEO_POLL_INTERVAL_SEC=10
+CUTVIDEO_MAX_WAIT_SEC=1800
+HUOBAO_BASE_URL=http://127.0.0.1:5678
+HUOBAO_POLL_INTERVAL_SEC=10
+HUOBAO_MAX_WAIT_SEC=900
 YTDLP_BIN=yt-dlp
 FFMPEG_BIN=ffmpeg
 WHISPER_BIN=
 WHISPER_MODEL=./data/collector/models/ggml-base.en.bin
-POSTGRES_DB=fastgpt
-POSTGRES_USER=postgres
-POSTGRES_PASSWORD=postgres
-MINIO_ROOT_USER=minioadmin
-MINIO_ROOT_PASSWORD=minioadmin
+N8N_IMAGE=docker.n8n.io/n8nio/n8n:latest
+WEBHOOK_URL=http://127.0.0.1:5670/
+GENERIC_TIMEZONE=Asia/Shanghai
+TZ=Asia/Shanghai
 CLIPROXY_IMAGE=storyforge/cli-proxy-api:patched
--- a/README.md
+++ b/README.md
@@ -5,36 +5,68 @@ StoryForge 现在拆成独立项目目录，和 `AI-glasses` 分开维护。
 ## 目录

 - `android-app/`：StoryForge Android 客户端
- `collector-service/`：FastAPI 后端，提供登录、审批、素材导入、知识库、智能体和 OTA
- `docker-compose.yml`：本地 FastGPT / collector / 基础依赖编排
+- `collector-service/`：FastAPI 后端，负责用户体系、项目、Agent、任务、内容分析和对外能力接入
+- `n8n/`：工作流导出文件，作为流程编排中枢
+- `docker-compose.yml`：本地 `collector + n8n + cli-proxy-api` 编排
 - `Common/`：项目约束和架构说明
 - `data/collector/`：SQLite、任务文件、下载产物
+- `docs/`：审计、实施计划、联调说明、当前 MVP 状态

 ## Android

 ```bash
-cd /Users/kris/code/StoryForge/android-app
+cd /Users/kris/code/StoryForge-gitea/android-app
 ./gradlew assembleDebug
 ```

 ## Collector Service

 ```bash
-cd /Users/kris/code/StoryForge/collector-service
+cd /Users/kris/code/StoryForge-gitea/collector-service
 python3 -m venv .venv
 source .venv/bin/activate
 pip install -r requirements.txt
 uvicorn app.main:app --host 0.0.0.0 --port 8081 --reload
 ```

+## Docker Compose
+
+```bash
+cd /Users/kris/code/StoryForge-gitea
+cp .env.example .env
+docker compose up -d --build
+```
+
+默认会启动：
+
+- `collector-service`：`http://127.0.0.1:8081`
+- `n8n`：`http://127.0.0.1:5670`
+- `cli-proxy-api`：`http://127.0.0.1:8317`
+
 默认会创建最高权限账号：

 - `kris`
 - `Asd123456.`

+## 当前架构
+
+- `collector-service` 负责：
+  - 用户账号、多项目、多 Agent、多任务、多内容源数据边界
+  - 调用下载器、本地 ASR、本机 OpenAI 兼容模型
+  - 调用 Windows `cutvideo` 和 `huobao-drama`
+  - 持久化任务、分镜、分析结果、事件日志
+- `n8n` 负责：
+  - 触发 `analysis_pipeline`
+  - 触发 `real_cut_pipeline`
+  - 触发 `ai_video_pipeline`
+- FastGPT 已从主流程设计中移除，不再作为运行时依赖
+
 ## 说明

 - 新注册账号默认 `pending`
 - 主管理员审批后才可使用核心业务接口
- 素材入口支持文字、视频链接、视频上传
- 可选对接本机 OpenAI 兼容模型服务和 FastGPT 数据集 API
+- 支持 `user -> project -> knowledge base / assistant(agent) / job / content source` 的多租户边界
+- 素材入口支持文字、视频链接、视频上传；内容源账号通过 `content_sources` 建模持久化
+- `cutvideo` 继续运行在 Windows 机器，本系统通过 API 调度
+- `huobao-drama` 继续作为 AI 生成视频主链的核心引擎
+- 详细审计、阶段计划和联调步骤见 `docs/`
--- a/android-app/app/src/main/java/com/aiglasses/app/storyforge/StoryForgeModels.kt
+++ b/android-app/app/src/main/java/com/aiglasses/app/storyforge/StoryForgeModels.kt
@@ -1,6 +1,10 @@
 package com.aiglasses.app.storyforge

 import kotlinx.serialization.Serializable
+import kotlinx.serialization.json.JsonArray
+import kotlinx.serialization.json.JsonObject
+import kotlinx.serialization.json.buildJsonArray
+import kotlinx.serialization.json.buildJsonObject

@Serializable
 data class RegisterAccountRequest(
@@ -66,12 +70,22 @@ data class PreferredModelRequest(
 )

@Serializable
-data class KnowledgeBaseDto(
+data class ProjectDto(
    val id: String,
    val user_id: String,
    val name: String,
    val description: String = "",
-    val fastgpt_dataset_id: String? = null,
+    val created_at: String = "",
+    val updated_at: String = ""
+)
+
+@Serializable
+data class KnowledgeBaseDto(
+    val id: String,
+    val user_id: String,
+    val project_id: String = "",
+    val name: String,
+    val description: String = "",
    val sync_status: String = "pending",
    val document_count: Int = 0,
    val linked_assistant_count: Int = 0,
@@ -82,6 +96,7 @@ data class KnowledgeBaseDto(
@Serializable
 data class KnowledgeBaseCreateRequest(
    val name: String,
+    val project_id: String = "",
    val description: String = ""
 )

@@ -89,12 +104,13 @@ data class KnowledgeBaseCreateRequest(
 data class AssistantDto(
    val id: String,
    val user_id: String,
+    val project_id: String = "",
    val name: String,
    val description: String = "",
    val system_prompt: String = "",
    val generation_goal: String = "",
    val knowledge_base_ids: List<String> = emptyList(),
-    val fastgpt_app_key: String = "",
+    val config: JsonObject = buildJsonObject { },
    val model_profile_id: String = "",
    val created_at: String = "",
    val updated_at: String = ""
@@ -107,7 +123,7 @@ data class AssistantCreateRequest(
    val system_prompt: String = "",
    val generation_goal: String = "",
    val knowledge_base_ids: List<String> = emptyList(),
-    val fastgpt_app_key: String = "",
+    val project_id: String = "",
    val model_profile_id: String = ""
 )

@@ -118,7 +134,7 @@ data class AssistantUpdateRequest(
    val system_prompt: String? = null,
    val generation_goal: String? = null,
    val knowledge_base_ids: List<String>? = null,
-    val fastgpt_app_key: String? = null,
+    val project_id: String? = null,
    val model_profile_id: String? = null
 )

@@ -126,6 +142,7 @@ data class AssistantUpdateRequest(
 data class ExploreVideoLinkRequest(
    val video_url: String,
    val title: String? = null,
+    val project_id: String? = null,
    val knowledge_base_id: String? = null,
    val assistant_id: String? = null,
    val analysis_model_profile_id: String? = null,
@@ -136,6 +153,7 @@ data class ExploreVideoLinkRequest(
 data class ExploreTextRequest(
    val title: String,
    val content: String,
+    val project_id: String? = null,
    val knowledge_base_id: String? = null,
    val assistant_id: String? = null,
    val analysis_model_profile_id: String? = null
@@ -145,19 +163,26 @@ data class ExploreTextRequest(
 data class JobDto(
    val id: String,
    val user_id: String,
+    val project_id: String = "",
    val assistant_id: String? = null,
    val knowledge_base_id: String,
+    val content_source_id: String = "",
    val source_type: String,
+    val line_type: String = "analysis",
+    val workflow_key: String = "",
+    val orchestrator: String = "n8n",
+    val provider_name: String = "",
+    val provider_task_id: String = "",
    val source_url: String? = null,
    val title: String,
    val language: String,
    val status: String,
    val transcript_text: String = "",
    val style_summary: String = "",
-    val fastgpt_collection_id: String = "",
    val upload_status: String = "pending",
    val error: String = "",
-    val artifacts: Map<String, String> = emptyMap(),
+    val artifacts: JsonObject = buildJsonObject { },
+    val result: JsonObject = buildJsonObject { },
    val analysis_model_profile_id: String = "",
    val created_at: String = "",
    val updated_at: String = ""
@@ -173,7 +198,9 @@ data class KnowledgeDocumentDto(
    val transcript_text: String = "",
    val style_summary: String = "",
    val combined_text: String = "",
-    val fastgpt_collection_id: String = "",
+    val analysis: JsonObject = buildJsonObject { },
+    val storyboards: JsonArray = buildJsonArray { },
+    val source_artifacts: JsonObject = buildJsonObject { },
    val analysis_model_profile_id: String = "",
    val created_at: String = "",
    val updated_at: String = ""
@@ -200,6 +227,7 @@ data class GenerateCopyResponseDto(
@Serializable
 data class DashboardDto(
    val account: AccountDto,
+    val projects: List<ProjectDto> = emptyList(),
    val knowledge_bases: List<KnowledgeBaseDto> = emptyList(),
    val assistants: List<AssistantDto> = emptyList(),
    val recent_jobs: List<JobDto> = emptyList(),
--- a/collector-service/app/database.py
+++ b/collector-service/app/database.py
@@ -48,6 +48,18 @@ class Database:
        with self.session() as conn:
            conn.execute(sql, params)

+    def table_exists(self, name: str) -> bool:
+        row = self.fetch_one(
+            "SELECT name FROM sqlite_master WHERE type = 'table' AND name = ?",
+            (name,),
+        )
+        return bool(row)
+
+    def column_exists(self, table: str, column: str) -> bool:
+        with self.session() as conn:
+            rows = conn.execute(f"PRAGMA table_info({table})").fetchall()
+        return any(row["name"] == column for row in rows)
+
    def init_schema(self) -> None:
        schema = """
        CREATE TABLE IF NOT EXISTS accounts (
@@ -90,10 +102,10 @@ class Database:
        CREATE TABLE IF NOT EXISTS knowledge_bases (
            id TEXT PRIMARY KEY,
            user_id TEXT NOT NULL,
+            project_id TEXT,
            name TEXT NOT NULL,
            description TEXT NOT NULL DEFAULT '',
-            fastgpt_dataset_id TEXT,
-            sync_status TEXT NOT NULL DEFAULT 'pending',
+            sync_status TEXT NOT NULL DEFAULT 'ready',
            created_at TEXT NOT NULL,
            updated_at TEXT NOT NULL,
            FOREIGN KEY(user_id) REFERENCES accounts(id) ON DELETE CASCADE
@@ -108,7 +120,9 @@ class Database:
            transcript_text TEXT NOT NULL DEFAULT '',
            style_summary TEXT NOT NULL DEFAULT '',
            combined_text TEXT NOT NULL DEFAULT '',
-            fastgpt_collection_id TEXT NOT NULL DEFAULT '',
+            analysis_json TEXT NOT NULL DEFAULT '{}',
+            storyboard_json TEXT NOT NULL DEFAULT '[]',
+            source_artifact_json TEXT NOT NULL DEFAULT '{}',
            analysis_model_profile_id TEXT NOT NULL DEFAULT '',
            created_at TEXT NOT NULL,
            updated_at TEXT NOT NULL,
@@ -118,11 +132,12 @@ class Database:
        CREATE TABLE IF NOT EXISTS assistants (
            id TEXT PRIMARY KEY,
            user_id TEXT NOT NULL,
+            project_id TEXT,
            name TEXT NOT NULL,
            description TEXT NOT NULL DEFAULT '',
            system_prompt TEXT NOT NULL DEFAULT '',
            generation_goal TEXT NOT NULL DEFAULT '',
-            fastgpt_app_key TEXT NOT NULL DEFAULT '',
+            config_json TEXT NOT NULL DEFAULT '{}',
            model_profile_id TEXT NOT NULL DEFAULT '',
            created_at TEXT NOT NULL,
            updated_at TEXT NOT NULL,
@@ -140,19 +155,26 @@ class Database:
        CREATE TABLE IF NOT EXISTS jobs (
            id TEXT PRIMARY KEY,
            user_id TEXT NOT NULL,
+            project_id TEXT,
            assistant_id TEXT,
            knowledge_base_id TEXT NOT NULL,
+            content_source_id TEXT,
            source_type TEXT NOT NULL,
+            line_type TEXT NOT NULL DEFAULT 'analysis',
+            workflow_key TEXT NOT NULL DEFAULT '',
+            orchestrator TEXT NOT NULL DEFAULT 'n8n',
+            provider_name TEXT NOT NULL DEFAULT '',
+            provider_task_id TEXT NOT NULL DEFAULT '',
            source_url TEXT,
            title TEXT NOT NULL,
            language TEXT NOT NULL DEFAULT 'auto',
            status TEXT NOT NULL,
            transcript_text TEXT NOT NULL DEFAULT '',
            style_summary TEXT NOT NULL DEFAULT '',
-            fastgpt_collection_id TEXT NOT NULL DEFAULT '',
            upload_status TEXT NOT NULL DEFAULT 'pending',
            error TEXT NOT NULL DEFAULT '',
            artifacts_json TEXT NOT NULL DEFAULT '{}',
+            result_json TEXT NOT NULL DEFAULT '{}',
            analysis_model_profile_id TEXT NOT NULL DEFAULT '',
            created_at TEXT NOT NULL,
            updated_at TEXT NOT NULL,
@@ -161,6 +183,42 @@ class Database:
            FOREIGN KEY(knowledge_base_id) REFERENCES knowledge_bases(id) ON DELETE CASCADE
        );

+        CREATE TABLE IF NOT EXISTS projects (
+            id TEXT PRIMARY KEY,
+            user_id TEXT NOT NULL,
+            name TEXT NOT NULL,
+            description TEXT NOT NULL DEFAULT '',
+            created_at TEXT NOT NULL,
+            updated_at TEXT NOT NULL,
+            FOREIGN KEY(user_id) REFERENCES accounts(id) ON DELETE CASCADE
+        );
+
+        CREATE TABLE IF NOT EXISTS content_sources (
+            id TEXT PRIMARY KEY,
+            user_id TEXT NOT NULL,
+            project_id TEXT,
+            source_kind TEXT NOT NULL,
+            platform TEXT NOT NULL DEFAULT '',
+            handle TEXT NOT NULL DEFAULT '',
+            source_url TEXT NOT NULL DEFAULT '',
+            title TEXT NOT NULL DEFAULT '',
+            local_path TEXT NOT NULL DEFAULT '',
+            metadata_json TEXT NOT NULL DEFAULT '{}',
+            created_at TEXT NOT NULL,
+            updated_at TEXT NOT NULL,
+            FOREIGN KEY(user_id) REFERENCES accounts(id) ON DELETE CASCADE,
+            FOREIGN KEY(project_id) REFERENCES projects(id) ON DELETE SET NULL
+        );
+
+        CREATE TABLE IF NOT EXISTS job_events (
+            id TEXT PRIMARY KEY,
+            job_id TEXT NOT NULL,
+            event_type TEXT NOT NULL,
+            payload_json TEXT NOT NULL DEFAULT '{}',
+            created_at TEXT NOT NULL,
+            FOREIGN KEY(job_id) REFERENCES jobs(id) ON DELETE CASCADE
+        );
+
        CREATE TABLE IF NOT EXISTS app_updates (
            id INTEGER PRIMARY KEY AUTOINCREMENT,
            platform TEXT NOT NULL,
@@ -179,3 +237,102 @@ class Database:
        """
        with self.session() as conn:
            conn.executescript(schema)
+        self.migrate_schema()
+
+    def migrate_schema(self) -> None:
+        table_columns: dict[str, dict[str, str]] = {
+            "knowledge_bases": {
+                "project_id": "TEXT",
+            },
+            "knowledge_documents": {
+                "analysis_json": "TEXT NOT NULL DEFAULT '{}'",
+                "storyboard_json": "TEXT NOT NULL DEFAULT '[]'",
+                "source_artifact_json": "TEXT NOT NULL DEFAULT '{}'",
+            },
+            "assistants": {
+                "project_id": "TEXT",
+                "config_json": "TEXT NOT NULL DEFAULT '{}'",
+            },
+            "jobs": {
+                "project_id": "TEXT",
+                "content_source_id": "TEXT",
+                "line_type": "TEXT NOT NULL DEFAULT 'analysis'",
+                "workflow_key": "TEXT NOT NULL DEFAULT ''",
+                "orchestrator": "TEXT NOT NULL DEFAULT 'n8n'",
+                "provider_name": "TEXT NOT NULL DEFAULT ''",
+                "provider_task_id": "TEXT NOT NULL DEFAULT ''",
+                "result_json": "TEXT NOT NULL DEFAULT '{}'",
+            },
+        }
+
+        for table, columns in table_columns.items():
+            if not self.table_exists(table):
+                continue
+            for column, definition in columns.items():
+                if self.column_exists(table, column):
+                    continue
+                self.execute(f"ALTER TABLE {table} ADD COLUMN {column} {definition}")
+
+        self.ensure_default_projects()
+
+    def ensure_default_projects(self) -> None:
+        if not self.table_exists("projects"):
+            return
+
+        accounts = self.fetch_all("SELECT id, username FROM accounts ORDER BY created_at ASC")
+        for account in accounts:
+            project = self.fetch_one(
+                "SELECT * FROM projects WHERE user_id = ? ORDER BY created_at ASC LIMIT 1",
+                (account["id"],),
+            )
+            if not project:
+                project_id = f"proj_{account['id']}"
+                now = utc_now()
+                self.execute(
+                    """
+                    INSERT INTO projects (id, user_id, name, description, created_at, updated_at)
+                    VALUES (?, ?, ?, ?, ?, ?)
+                    """,
+                    (
+                        project_id,
+                        account["id"],
+                        f"{account['username']} 默认项目",
+                        "系统自动创建的默认项目",
+                        now,
+                        now,
+                    ),
+                )
+                project = self.fetch_one("SELECT * FROM projects WHERE id = ?", (project_id,))
+
+            if not project:
+                continue
+
+            if self.column_exists("knowledge_bases", "project_id"):
+                self.execute(
+                    """
+                    UPDATE knowledge_bases
+                    SET project_id = ?
+                    WHERE user_id = ? AND (project_id IS NULL OR project_id = '')
+                    """,
+                    (project["id"], account["id"]),
+                )
+
+            if self.column_exists("assistants", "project_id"):
+                self.execute(
+                    """
+                    UPDATE assistants
+                    SET project_id = ?
+                    WHERE user_id = ? AND (project_id IS NULL OR project_id = '')
+                    """,
+                    (project["id"], account["id"]),
+                )
+
+            if self.column_exists("jobs", "project_id"):
+                self.execute(
+                    """
+                    UPDATE jobs
+                    SET project_id = ?
+                    WHERE user_id = ? AND (project_id IS NULL OR project_id = '')
+                    """,
+                    (project["id"], account["id"]),
+                )
--- a/collector-service/app/fastgpt.py
+++ b/collector-service/app/fastgpt.py
@@ -1,48 +0,0 @@
-from __future__ import annotations
-
-from typing import Any
-
-import httpx
-
-
-class FastGPTClient:
-    def __init__(self, *, base_url: str, dataset_api_key: str, timeout: float = 60.0) -> None:
-        self.base_url = base_url.rstrip("/")
-        self.dataset_api_key = dataset_api_key.strip()
-        self.timeout = timeout
-
-    @property
-    def enabled(self) -> bool:
-        return bool(self.base_url and self.dataset_api_key)
-
-    async def ensure_dataset(self, name: str, intro: str = "") -> dict[str, Any] | None:
-        if not self.enabled:
-            return None
-        payload = {"name": name, "intro": intro}
-        async with httpx.AsyncClient(timeout=self.timeout) as client:
-            response = await client.post(
-                f"{self.base_url}/api/core/dataset/create",
-                headers={"Authorization": f"Bearer {self.dataset_api_key}"},
-                json=payload,
-            )
-            response.raise_for_status()
-            return response.json().get("data") or response.json()
-
-    async def add_text_document(self, dataset_id: str, name: str, text: str) -> dict[str, Any] | None:
-        if not self.enabled or not dataset_id.strip():
-            return None
-        payload = {
-            "datasetId": dataset_id,
-            "type": "text",
-            "name": name,
-            "trainingType": "chunk",
-            "text": text,
-        }
-        async with httpx.AsyncClient(timeout=self.timeout) as client:
-            response = await client.post(
-                f"{self.base_url}/api/core/dataset/collection/create/text",
-                headers={"Authorization": f"Bearer {self.dataset_api_key}"},
-                json=payload,
-            )
-            response.raise_for_status()
-            return response.json().get("data") or response.json()
--- a/collector-service/app/integrations.py
+++ b/collector-service/app/integrations.py
@@ -0,0 +1,162 @@
+from __future__ import annotations
+
+from typing import Any
+
+import httpx
+
+
+def _join_url(base_url: str, path: str) -> str:
+    base = base_url.rstrip("/")
+    if path.startswith("http://") or path.startswith("https://"):
+        return path
+    return f"{base}/{path.lstrip('/')}"
+
+
+def _unwrap_response(payload: Any) -> dict[str, Any]:
+    if not isinstance(payload, dict):
+        return {"value": payload}
+    if payload.get("success") is True and "data" in payload:
+        data = payload.get("data")
+        if isinstance(data, dict):
+            return data
+        return {"value": data}
+    return payload
+
+
+class N8NClient:
+    def __init__(
+        self,
+        *,
+        base_url: str,
+        workflow_paths: dict[str, str],
+        shared_secret: str = "",
+        timeout: float = 60.0,
+    ) -> None:
+        self.base_url = base_url.rstrip("/")
+        self.workflow_paths = workflow_paths
+        self.shared_secret = shared_secret.strip()
+        self.timeout = timeout
+
+    @property
+    def enabled(self) -> bool:
+        return bool(self.base_url)
+
+    async def trigger(self, workflow_key: str, payload: dict[str, Any]) -> dict[str, Any]:
+        workflow_path = self.workflow_paths.get(workflow_key, "").strip()
+        if not workflow_path:
+            raise ValueError(f"workflow path not configured for {workflow_key}")
+        try:
+            workflow_path = workflow_path.format(**payload)
+        except KeyError:
+            pass
+        headers: dict[str, str] = {}
+        if self.shared_secret:
+            headers["X-Orchestrator-Secret"] = self.shared_secret
+        async with httpx.AsyncClient(timeout=self.timeout) as client:
+            response = await client.post(
+                _join_url(self.base_url, workflow_path),
+                json=payload,
+                headers=headers,
+            )
+            response.raise_for_status()
+            if not response.content:
+                return {"triggered": True}
+            return _unwrap_response(response.json())
+
+
+class CutVideoClient:
+    def __init__(self, *, base_url: str, api_key: str = "", timeout: float = 120.0) -> None:
+        self.base_url = base_url.rstrip("/")
+        self.api_key = api_key.strip()
+        self.timeout = timeout
+
+    @property
+    def enabled(self) -> bool:
+        return bool(self.base_url)
+
+    def _headers(self) -> dict[str, str]:
+        headers: dict[str, str] = {}
+        if self.api_key:
+            headers["Authorization"] = f"Bearer {self.api_key}"
+        return headers
+
+    async def submit_job(self, payload: dict[str, Any]) -> dict[str, Any]:
+        async with httpx.AsyncClient(timeout=self.timeout) as client:
+            response = await client.post(
+                _join_url(self.base_url, "/api/jobs"),
+                json=payload,
+                headers=self._headers(),
+            )
+            response.raise_for_status()
+            return _unwrap_response(response.json())
+
+    async def get_task(self, task_id: str) -> dict[str, Any]:
+        async with httpx.AsyncClient(timeout=self.timeout) as client:
+            response = await client.get(
+                _join_url(self.base_url, f"/api/tasks/{task_id}"),
+                headers=self._headers(),
+            )
+            response.raise_for_status()
+            return _unwrap_response(response.json())
+
+    async def get_run(self, run_id: str) -> dict[str, Any]:
+        async with httpx.AsyncClient(timeout=self.timeout) as client:
+            response = await client.get(
+                _join_url(self.base_url, f"/api/runs/{run_id}"),
+                headers=self._headers(),
+            )
+            response.raise_for_status()
+            return _unwrap_response(response.json())
+
+
+class HuobaoDramaClient:
+    def __init__(self, *, base_url: str, timeout: float = 180.0) -> None:
+        self.base_url = base_url.rstrip("/")
+        self.timeout = timeout
+
+    @property
+    def enabled(self) -> bool:
+        return bool(self.base_url)
+
+    async def create_drama(self, payload: dict[str, Any]) -> dict[str, Any]:
+        async with httpx.AsyncClient(timeout=self.timeout) as client:
+            response = await client.post(
+                _join_url(self.base_url, "/api/v1/dramas"),
+                json=payload,
+            )
+            response.raise_for_status()
+            return _unwrap_response(response.json())
+
+    async def generate_image(self, payload: dict[str, Any]) -> dict[str, Any]:
+        async with httpx.AsyncClient(timeout=self.timeout) as client:
+            response = await client.post(
+                _join_url(self.base_url, "/api/v1/images"),
+                json=payload,
+            )
+            response.raise_for_status()
+            return _unwrap_response(response.json())
+
+    async def get_image(self, image_id: str) -> dict[str, Any]:
+        async with httpx.AsyncClient(timeout=self.timeout) as client:
+            response = await client.get(
+                _join_url(self.base_url, f"/api/v1/images/{image_id}"),
+            )
+            response.raise_for_status()
+            return _unwrap_response(response.json())
+
+    async def generate_video(self, payload: dict[str, Any]) -> dict[str, Any]:
+        async with httpx.AsyncClient(timeout=self.timeout) as client:
+            response = await client.post(
+                _join_url(self.base_url, "/api/v1/videos"),
+                json=payload,
+            )
+            response.raise_for_status()
+            return _unwrap_response(response.json())
+
+    async def get_video(self, video_id: str) -> dict[str, Any]:
+        async with httpx.AsyncClient(timeout=self.timeout) as client:
+            response = await client.get(
+                _join_url(self.base_url, f"/api/v1/videos/{video_id}"),
+            )
+            response.raise_for_status()
+            return _unwrap_response(response.json())
--- a/collector-service/app/main.py
+++ b/collector-service/app/main.py
--- a/docker-compose.yml
+++ b/docker-compose.yml
@@ -1,56 +1,30 @@
-version: "3.9"
-
 services:
-  mongo:
-    image: mongo:6
-    container_name: storyforge-mongo
-    restart: unless-stopped
-    ports:
-      - "27017:27017"
-    volumes:
-      - ./data/mongo:/data/db
-
-  vectorDB:
-    image: pgvector/pgvector:pg16
-    container_name: storyforge-pgvector
+  n8n:
+    image: ${N8N_IMAGE:-docker.n8n.io/n8nio/n8n:latest}
+    container_name: storyforge-n8n
    restart: unless-stopped
    environment:
-      POSTGRES_DB: ${POSTGRES_DB:-fastgpt}
-      POSTGRES_USER: ${POSTGRES_USER:-postgres}
-      POSTGRES_PASSWORD: ${POSTGRES_PASSWORD:-postgres}
+      N8N_HOST: ${N8N_HOST:-0.0.0.0}
+      N8N_PORT: 5678
+      N8N_PROTOCOL: ${N8N_PROTOCOL:-http}
+      WEBHOOK_URL: ${WEBHOOK_URL:-http://127.0.0.1:5670/}
+      GENERIC_TIMEZONE: ${GENERIC_TIMEZONE:-Asia/Shanghai}
+      TZ: ${TZ:-Asia/Shanghai}
+      N8N_SECURE_COOKIE: ${N8N_SECURE_COOKIE:-false}
+      N8N_ENFORCE_SETTINGS_FILE_PERMISSIONS: ${N8N_ENFORCE_SETTINGS_FILE_PERMISSIONS:-false}
    ports:
-      - "5432:5432"
+      - "5670:5678"
    volumes:
-      - ./data/pg:/var/lib/postgresql/data
-
-  redis:
-    image: redis:7-alpine
-    container_name: storyforge-redis
-    restart: unless-stopped
-    ports:
-      - "6379:6379"
-    volumes:
-      - ./data/redis:/data
-
-  minio:
-    image: minio/minio:RELEASE.2025-02-07T23-21-09Z
-    container_name: storyforge-minio
-    restart: unless-stopped
-    command: server /data --console-address ":9001"
-    environment:
-      MINIO_ROOT_USER: ${MINIO_ROOT_USER:-minioadmin}
-      MINIO_ROOT_PASSWORD: ${MINIO_ROOT_PASSWORD:-minioadmin}
-    ports:
-      - "9000:9000"
-      - "9001:9001"
-    volumes:
-      - ./data/minio:/data
+      - ./data/n8n:/home/node/.n8n
+      - ./n8n:/workspace/n8n:ro

  collector:
    build:
      context: ./collector-service
    container_name: storyforge-collector
    restart: unless-stopped
+    depends_on:
+      - n8n
    environment:
      DATA_DIR: /data/collector
      DATABASE_PATH: /data/collector/storyforge.db
@@ -58,40 +32,29 @@ services:
      LOCAL_OPENAI_BASE_URL: ${LOCAL_OPENAI_BASE_URL:-http://host.docker.internal:8317/v1}
      LOCAL_OPENAI_MODEL: ${LOCAL_OPENAI_MODEL:-GLM-5}
      LOCAL_OPENAI_API_KEY: ${LOCAL_OPENAI_API_KEY:-}
-      FASTGPT_BASE_URL: ${FASTGPT_BASE_URL:-http://host.docker.internal:3000}
-      FASTGPT_DATASET_API_KEY: ${FASTGPT_DATASET_API_KEY:-}
+      N8N_BASE_URL: ${N8N_BASE_URL:-http://n8n:5678}
+      N8N_ANALYSIS_WEBHOOK_PATH: ${N8N_ANALYSIS_WEBHOOK_PATH:-/webhook/storyforge-analysis}
+      N8N_REAL_CUT_WEBHOOK_PATH: ${N8N_REAL_CUT_WEBHOOK_PATH:-/webhook/storyforge-real-cut}
+      N8N_AI_VIDEO_WEBHOOK_PATH: ${N8N_AI_VIDEO_WEBHOOK_PATH:-/webhook/storyforge-ai-video}
+      ORCHESTRATOR_SHARED_SECRET: ${ORCHESTRATOR_SHARED_SECRET:-storyforge-local-secret}
+      CUTVIDEO_BASE_URL: ${CUTVIDEO_BASE_URL:-}
+      CUTVIDEO_API_KEY: ${CUTVIDEO_API_KEY:-}
+      CUTVIDEO_BASE_CONFIG: ${CUTVIDEO_BASE_CONFIG:-example.job.yaml}
+      CUTVIDEO_POLL_INTERVAL_SEC: ${CUTVIDEO_POLL_INTERVAL_SEC:-10}
+      CUTVIDEO_MAX_WAIT_SEC: ${CUTVIDEO_MAX_WAIT_SEC:-1800}
+      HUOBAO_BASE_URL: ${HUOBAO_BASE_URL:-http://host.docker.internal:5678}
      YTDLP_BIN: ${YTDLP_BIN:-yt-dlp}
      FFMPEG_BIN: ${FFMPEG_BIN:-ffmpeg}
      WHISPER_BIN: ${WHISPER_BIN:-}
      WHISPER_MODEL: ${WHISPER_MODEL:-/data/collector/models/ggml-base.en.bin}
+      HUOBAO_POLL_INTERVAL_SEC: ${HUOBAO_POLL_INTERVAL_SEC:-10}
+      HUOBAO_MAX_WAIT_SEC: ${HUOBAO_MAX_WAIT_SEC:-900}
    ports:
      - "8081:8081"
    volumes:
      - ./data/collector:/data/collector
    command: uvicorn app.main:app --host 0.0.0.0 --port 8081

-  fastgpt:
-    image: ghcr.io/labring/fastgpt:latest
-    container_name: storyforge-fastgpt
-    restart: unless-stopped
-    depends_on:
-      - mongo
-      - vectorDB
-      - redis
-      - minio
-    ports:
-      - "3000:3000"
-
-  sandbox:
-    image: ghcr.io/labring/fastgpt-sandbox:latest
-    container_name: storyforge-sandbox
-    restart: unless-stopped
-
-  fastgpt-plugin:
-    image: ghcr.io/labring/fastgpt-plugin:latest
-    container_name: storyforge-fastgpt-plugin
-    restart: unless-stopped
-
  cli-proxy-api:
    image: ${CLIPROXY_IMAGE:-storyforge/cli-proxy-api:patched}
    container_name: storyforge-cliproxyapi
--- a/docs/AUDIT_2026-03-18.md
+++ b/docs/AUDIT_2026-03-18.md
@@ -0,0 +1,134 @@
+# StoryForge 现状审计
+
+日期：2026-03-18
+
+## 结论
+
+当前应以 `/Users/kris/code/StoryForge-gitea` 作为主工作区继续推进，而不是 `/Users/kris/code/Fastgpt`。后者更像一次不完整的导入快照，前者才是可持续开发的真实仓库。
+
+## 现有功能归位
+
+### 1. `collector-service` 之前承担的功能
+
+- 账号注册、登录、审批
+- 本地模型配置
+- 知识库、智能体、任务管理
+- 视频链接/上传视频/文本三类入口
+- 下载器、ffmpeg、whisper.cpp 风格的本地处理调用
+- Android OTA 查询/发布
+
+### 2. FastGPT 实际承担的功能
+
+- 仅承担“数据集/文档同步”的外部依赖角色
+- 代码痕迹集中在：
+  - `collector-service/app/fastgpt.py`
+  - `docker-compose.yml`
+  - 若干 `fastgpt_*` 字段
+
+结论：FastGPT 并不是业务内核，适合迁移后整体删除。
+
+### 3. n8n 适合接管的功能
+
+- 任务触发
+- 工作流分流
+- 外部能力编排入口
+- 任务执行顺序控制
+
+不适合承载：
+
+- 用户、项目、Agent、知识库、任务、历史记录的主数据
+- 业务状态唯一真相源
+
+结论：应采用“业务状态在 `collector-service`，流程编排在 `n8n`”的分层。
+
+## 多用户与数据边界
+
+当前已明确采用：
+
+- `accounts`
+- `projects`
+- `knowledge_bases`
+- `assistants`
+- `content_sources`
+- `jobs`
+- `job_events`
+
+推荐模型：`user + project`。
+
+理由：
+
+- 只做 `user` 级隔离，会导致一个用户内部不同内容工作流难以再分边界
+- `project` 可以自然承接“一个创作者方向 / 一个客户 / 一个账号矩阵 / 一个内容实验”
+- `assistant`、`knowledge_base`、`job`、`content_source` 都能挂到 `project`，便于后续扩展协作空间
+
+## 外部链路审计
+
+### 1. 下载器
+
+- 已存在，不需要重写
+- 现阶段通过 `yt-dlp` 命令集成
+
+### 2. ASR
+
+- 现有实现已部署，但入口未完全标准化
+- 当前后端支持 `ffmpeg + whisper.cpp` 风格接入
+- 若需要接现有常驻 ASR 服务，后续只需把 `transcribe_media()` 改成 HTTP/RPC 适配即可
+
+### 3. Windows `cutvideo`
+
+- 仓库：`/Users/kris/code/cutvideo`
+- 具备清晰 API：
+  - `POST /api/jobs`
+  - `GET /api/tasks/{task_id}`
+  - `GET /api/runs/{run_id}`
+- 适合集成为“由 StoryForge 后端授权调用的局域网剪辑能力”
+
+当前限制：
+
+- 现有 `cutvideo` API 主要接受 `input_dir`
+- 对“用户上传实拍素材后直接推送到 Windows 机器”这一步，还缺一层文件转运方案
+
+### 4. `huobao-drama`
+
+- 旧改版位置：`/Users/kris/code/huobaoduanju/huobao-drama-master`
+- 最新 upstream：`/Users/kris/code/huobao-drama-upstream`
+- 旧改版主要多了一套 `ad_workflow` 方向，和当前 StoryForge MVP 不完全对齐
+- 最新版已具备：
+  - `POST /api/v1/dramas`
+  - `POST /api/v1/images`
+  - `GET /api/v1/images/{id}`
+  - `POST /api/v1/videos`
+  - `GET /api/v1/videos/{id}`
+  - `reference_mode=first_last`
+
+本次真实联调里，旧改版为了兼容 `qnaigc` 需要补 4 个点：
+
+- `pkg/image/openai_image_client.go`
+- `application/services/image_generation_service.go`
+- `pkg/video/openai_sora_client.go`
+- `application/services/video_generation_service.go`
+
+核对结果：
+
+- 以上 4 个文件与本机 upstream 同名文件在补丁前没有明显结构分叉
+- 当前差异基本就是 `qnaigc` 图片异步查询、Kling 视频 JSON 协议、结果 URL 解析、远程首尾帧 URL 保留这几处兼容逻辑
+
+结论：这批补丁是可移植补丁，MVP 已在旧改版实例上验证通过；下一步应把同样补丁迁到最新版 `huobao-drama-upstream`，而不是继续在旧目录长期演进。
+
+## 当前已完成迁移面
+
+- FastGPT 运行时依赖已从 `collector-service` 主代码中剥离
+- 数据库已支持 `project/content_source/job_events`
+- `collector-service` 已增加：
+  - `n8n` 触发
+  - `cutvideo` 集成 client
+  - `huobao-drama` 集成 client
+  - 内部编排接口
+- `docker-compose.yml` 已改为 `collector + n8n + cli-proxy-api`
+- `n8n` 工作流导出文件已纳入仓库
+
+## 当前主要风险
+
+1. `cutvideo` 的素材传输还未完整闭环
+2. 本地 ASR 的“最终生产入口”仍需按你现有部署方式再绑一次
+3. `huobao-drama` 已在本机旧改版实例上跑通，但兼容补丁尚未迁到 upstream 仓库并形成正式提交
--- a/docs/IMPLEMENTATION_PLAN_2026-03-18.md
+++ b/docs/IMPLEMENTATION_PLAN_2026-03-18.md
@@ -0,0 +1,97 @@
+# StoryForge 分阶段实施计划
+
+日期：2026-03-18
+
+## Phase 0: 审计与基线收拢
+
+- 确认主工作区
+- 识别 FastGPT 真实职责
+- 识别多用户、多项目需要的主数据模型
+- 对比 `huobao-drama` 旧改版与 upstream
+- 审计 `cutvideo` 接口能力
+
+状态：已完成
+
+## Phase 1: 业务后端改造成主状态中心
+
+- 引入 `projects`
+- 引入 `content_sources`
+- 引入 `job_events`
+- 让 `knowledge_bases / assistants / jobs` 全部 project 化
+- 去掉 `collector-service` 中的 FastGPT 运行时逻辑
+- 增加 `agents` 别名接口，统一 Agent 语义
+
+状态：已完成首版
+
+## Phase 2: n8n 接管流程编排
+
+- 公共任务创建接口只负责建任务并触发工作流
+- `n8n` 负责分发：
+  - `analysis_pipeline`
+  - `real_cut_pipeline`
+  - `ai_video_pipeline`
+- 业务步骤落在 `collector-service` 内部接口，保证状态统一入库
+
+状态：已完成首版
+
+## Phase 3: 内容分析主线 MVP
+
+- 支持文本
+- 支持视频链接
+- 支持上传视频
+- 接下载器
+- 接本地 ASR
+- 接本地 LLM
+- 产出：
+  - transcript
+  - style_summary
+  - analysis
+  - rewrite
+  - storyboards
+
+状态：已完成首版
+
+## Phase 4: 实拍自动剪辑主线 MVP
+
+- 建立 `real_cut` 任务类型
+- 通过 `n8n -> collector -> cutvideo` 调度 Windows 机器
+- 记录 `task_id / run_id / 结果产物`
+
+状态：已完成 API 级集成
+
+待补：
+
+- 用户上传素材到 Windows 侧的文件转运闭环
+
+## Phase 5: AI 自动生成视频主线 MVP
+
+- 建立 `ai_video` 任务类型
+- 从分析结果或直接 brief 生成分镜
+- 调 `huobao-drama`：
+  - 创建 drama
+  - 生成首帧
+  - 生成尾帧
+  - 基于首尾帧生成视频
+- 结果回写任务
+
+状态：已完成 API 级集成
+
+## Phase 6: 删除 FastGPT 运行依赖
+
+- 删除代码依赖
+- 删除 compose 服务
+- 删除环境变量
+- 删除 README 说明
+
+状态：已完成主仓库首版
+
+## Phase 7: 联调与验证
+
+- Python 语法检查
+- Compose 配置检查
+- `collector-service` 本地启动
+- `n8n` workflow 导入
+- Windows `cutvideo` 局域网调度
+- `huobao-drama` 本机调用
+
+状态：进行中
--- a/docs/LAN_E2E_GUIDE_2026-03-18.md
+++ b/docs/LAN_E2E_GUIDE_2026-03-18.md
@@ -0,0 +1,146 @@
+# StoryForge 本地 / 局域网联调说明
+
+日期：2026-03-18
+
+## 1. 准备 `.env`
+
+复制：
+
+```bash
+cd /Users/kris/code/StoryForge-gitea
+cp .env.example .env
+```
+
+至少确认这些变量：
+
+- `N8N_BASE_URL=http://127.0.0.1:5670`
+- `ORCHESTRATOR_SHARED_SECRET=storyforge-local-secret`
+- `CUTVIDEO_BASE_URL=http://<windows-lan-ip>:7860`
+- `CUTVIDEO_API_KEY=` 如果 Windows 服务启用了鉴权
+- `HUOBAO_BASE_URL=http://127.0.0.1:5678`
+- `WHISPER_BIN=` 指向你现有本地 ASR 可执行文件时填写
+
+说明：
+
+- 如果你单独重建 `collector`，要确保运行时仍带上 `CUTVIDEO_BASE_URL`，否则容器会退回空值
+- 当前已验证可用的 Windows `cutvideo` 地址是 `http://192.168.31.18:7860`
+
+## 2. 启动基础服务
+
+```bash
+cd /Users/kris/code/StoryForge-gitea
+docker compose up -d --build
+```
+
+检查：
+
+- `collector-service`：`http://127.0.0.1:8081/healthz`
+- `n8n`：`http://127.0.0.1:5670`
+- `cli-proxy-api`：`http://127.0.0.1:8317`
+- 本机 `huobao-drama`：`http://127.0.0.1:5678/health`
+
+## 3. 导入 n8n workflows
+
+从 `n8n/workflows/` 导入：
+
+- `storyforge-analysis.json`
+- `storyforge-real-cut.json`
+- `storyforge-ai-video.json`
+
+导入后：
+
+- 检查每个 HTTP Request 节点的 `X-Orchestrator-Secret`
+- 如果你改了 `.env` 的 secret，这里必须同步
+
+## 4. 登录与审批
+
+默认超级管理员：
+
+- 用户名：`kris`
+- 密码：`Asd123456.`
+
+新用户注册后，需要用超级管理员审批。
+
+## 5. 内容分析链路验证
+
+### 文本
+
+调用 `POST /v2/explore/text`
+
+预期：
+
+- 任务创建成功
+- `n8n` webhook 被触发
+- 任务最终进入 `completed`
+- 知识库文档里出现 transcript / style_summary / analysis / storyboards
+
+已验证样例：
+
+- `job_203bc8e9b20f4b1cbbc6cf7da79e46f4`
+
+### 视频链接
+
+调用 `POST /v2/explore/video-link`
+
+前提：
+
+- `yt-dlp` 可用
+- `ffmpeg` 可用
+- ASR 可调用
+
+### 上传视频
+
+调用 `POST /v2/explore/upload-video`
+
+预期与视频链接类似，但素材来源为本地上传
+
+## 6. `cutvideo` 实拍剪辑链路验证
+
+调用 `POST /v2/pipelines/real-cut`
+
+当前 MVP 前提：
+
+- `input_dir` 必须是 Windows `cutvideo` 机器可访问的目录
+- 该目录中的素材已准备好
+
+预期：
+
+- 任务创建成功
+- `n8n` 调用 `collector-service` 内部 real-cut step
+- 后端记录 `provider_task_id`
+- 最终任务写回 `cutvideo_run`
+
+已验证样例：
+
+- `job_5ebd829c3f2144bca5c941183e75bdcd`
+- Windows 返回 `task_id=8d8f4a0cd5d9`
+- 运行目录 `20260318-093520-Windows cutvideo 联调样例`
+
+## 7. `huobao-drama` AI 视频链路验证
+
+调用 `POST /v2/pipelines/ai-video`
+
+推荐方式：
+
+- 先完成一个分析任务
+- 再把该分析任务的 `source_job_id` 传给 AI 视频任务
+
+预期：
+
+- 创建 drama
+- 每个分镜生成首帧、尾帧
+- 每个分镜生成视频
+- 最终 `job.result.rendered_scenes` 有完整结果
+
+已验证样例：
+
+- `job_01828c40377747cf914b51be360cc333`
+- `provider_task_id=10`
+- `video.task_id=qvideo-1380265978-1773799215825814468`
+- 最终视频已回写到 `job.result.rendered_scenes[0].video.video_url`
+
+## 8. 当前已知卡点
+
+- `cutvideo` 端到端“上传素材后自动送到 Windows 机器”还未彻底闭环
+- ASR 如果不是命令行模式，而是你现有常驻服务模式，需要再做一次入口绑定
+- `huobao-drama` 目前跑通依赖本地旧改版中的 qnaigc 兼容补丁，下一步要迁到 upstream 仓库
--- a/docs/MVP_STATUS_2026-03-18.md
+++ b/docs/MVP_STATUS_2026-03-18.md
@@ -0,0 +1,41 @@
+# StoryForge MVP 状态
+
+日期：2026-03-18
+
+## 已跑通或已完成代码接通
+
+- 多用户账号体系
+- 审批机制
+- `user -> project -> assistant / knowledge base / job / content source` 数据模型
+- 文本 / 视频链接 / 上传视频 三类分析任务创建
+- `n8n` 工作流导入、激活与触发接口
+- 本地下载器调用
+- 本地 `ffmpeg` / `whisper` 风格入口封装
+- 本地大模型内容分析、二创文案、分镜生成
+- Windows `cutvideo` API 调度与结果回写接口
+- 本机 `huobao-drama` API 调度、首尾帧生成、视频生成与结果回写接口
+- FastGPT 运行时依赖删除
+
+## 已验证的真实任务
+
+- 分析链路：`job_203bc8e9b20f4b1cbbc6cf7da79e46f4`
+- 实拍剪辑链路：`job_5ebd829c3f2144bca5c941183e75bdcd`
+- AI 视频链路：`job_01828c40377747cf914b51be360cc333`
+
+## 已实现但仍待环境验证
+
+- 现有 ASR 部署入口与 `collector-service` 的最终绑定
+
+## 尚未完全跑通
+
+- 用户上传实拍素材后，自动把素材转运到 Windows `cutvideo` 机器的闭环
+- 对“抖音 / bilibili / 小红书账号级内容源”的批量抓取与分析调度
+- `huobao-drama` 本地兼容补丁向 upstream 仓库的迁移、分支化和提交
+
+## 下一步优先级
+
+1. 把 `huobao-drama` 本地兼容补丁迁到 `/Users/kris/code/huobao-drama-upstream`
+2. 绑定你的真实 ASR 入口
+3. 决定实拍素材转运方案：共享目录优先，上传 API 作为备选
+4. 补账号级内容源抓取调度
+5. 把改动整理成提交并推送
--- a/n8n/README.md
+++ b/n8n/README.md
@@ -0,0 +1,28 @@
+# n8n Workflows
+
+本目录保存 StoryForge 的工作流导出文件，避免流程只存在于 n8n UI。
+
+## 工作流
+
+- `workflows/storyforge-analysis.json`：内容分析主线
+- `workflows/storyforge-real-cut.json`：Windows `cutvideo` 调度主线
+- `workflows/storyforge-ai-video.json`：`huobao-drama` AI 生成视频主线
+
+## 约定
+
+- 工作流内部默认通过 `http://collector:8081` 调用 `collector-service`
+- 内部调用头部使用 `X-Orchestrator-Secret: storyforge-local-secret`
+- 如果你修改了 `.env` 里的 `ORCHESTRATOR_SHARED_SECRET`，导入工作流后需要同步更新对应 HTTP Request 节点
+
+## 导入
+
+1. 先执行 `docker compose up -d n8n collector`
+2. 打开 `http://127.0.0.1:5670`
+3. 从 UI 导入本目录下的 3 个 JSON
+4. 激活工作流
+
+## Webhook 路径
+
+- `/webhook/storyforge-analysis`
+- `/webhook/storyforge-real-cut`
+- `/webhook/storyforge-ai-video`
--- a/n8n/workflows/storyforge-ai-video.json
+++ b/n8n/workflows/storyforge-ai-video.json
@@ -0,0 +1,70 @@
+{
+  "name": "StoryForge AI Video Pipeline",
+  "nodes": [
+    {
+      "parameters": {
+        "httpMethod": "POST",
+        "path": "storyforge-ai-video",
+        "responseMode": "onReceived",
+        "options": {}
+      },
+      "id": "aivideo-webhook",
+      "name": "AI Video Webhook",
+      "type": "n8n-nodes-base.webhook",
+      "typeVersion": 2,
+      "position": [
+        220,
+        300
+      ],
+      "webhookId": "storyforge-ai-video"
+    },
+    {
+      "parameters": {
+        "method": "POST",
+        "url": "={{'http://collector:8081/internal/jobs/steps/ai-video/render?job_id=' + ($json.body.job_id || $json.body.jobId)}}",
+        "sendHeaders": true,
+        "headerParameters": {
+          "parameters": [
+            {
+              "name": "X-Orchestrator-Secret",
+              "value": "storyforge-local-secret"
+            }
+          ]
+        },
+        "options": {
+          "timeout": 3600000
+        }
+      },
+      "id": "aivideo-runner",
+      "name": "Run AI Video Step",
+      "type": "n8n-nodes-base.httpRequest",
+      "typeVersion": 4.2,
+      "position": [
+        520,
+        300
+      ]
+    }
+  ],
+  "connections": {
+    "AI Video Webhook": {
+      "main": [
+        [
+          {
+            "node": "Run AI Video Step",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Run AI Video Step": {
+      "main": [
+        []
+      ]
+    }
+  },
+  "active": false,
+  "settings": {},
+  "pinData": {},
+  "versionId": "storyforge-ai-video-v1"
+}
--- a/n8n/workflows/storyforge-analysis.json
+++ b/n8n/workflows/storyforge-analysis.json
@@ -0,0 +1,70 @@
+{
+  "name": "StoryForge Analysis Pipeline",
+  "nodes": [
+    {
+      "parameters": {
+        "httpMethod": "POST",
+        "path": "storyforge-analysis",
+        "responseMode": "onReceived",
+        "options": {}
+      },
+      "id": "analysis-webhook",
+      "name": "Analysis Webhook",
+      "type": "n8n-nodes-base.webhook",
+      "typeVersion": 2,
+      "position": [
+        220,
+        300
+      ],
+      "webhookId": "storyforge-analysis"
+    },
+    {
+      "parameters": {
+        "method": "POST",
+        "url": "={{'http://collector:8081/internal/jobs/steps/analyze?job_id=' + ($json.body.job_id || $json.body.jobId)}}",
+        "sendHeaders": true,
+        "headerParameters": {
+          "parameters": [
+            {
+              "name": "X-Orchestrator-Secret",
+              "value": "storyforge-local-secret"
+            }
+          ]
+        },
+        "options": {
+          "timeout": 600000
+        }
+      },
+      "id": "analysis-runner",
+      "name": "Run Analysis Step",
+      "type": "n8n-nodes-base.httpRequest",
+      "typeVersion": 4.2,
+      "position": [
+        520,
+        300
+      ]
+    }
+  ],
+  "connections": {
+    "Analysis Webhook": {
+      "main": [
+        [
+          {
+            "node": "Run Analysis Step",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Run Analysis Step": {
+      "main": [
+        []
+      ]
+    }
+  },
+  "active": false,
+  "settings": {},
+  "pinData": {},
+  "versionId": "storyforge-analysis-v1"
+}
--- a/n8n/workflows/storyforge-real-cut.json
+++ b/n8n/workflows/storyforge-real-cut.json
@@ -0,0 +1,70 @@
+{
+  "name": "StoryForge Real Cut Pipeline",
+  "nodes": [
+    {
+      "parameters": {
+        "httpMethod": "POST",
+        "path": "storyforge-real-cut",
+        "responseMode": "onReceived",
+        "options": {}
+      },
+      "id": "realcut-webhook",
+      "name": "Real Cut Webhook",
+      "type": "n8n-nodes-base.webhook",
+      "typeVersion": 2,
+      "position": [
+        220,
+        300
+      ],
+      "webhookId": "storyforge-real-cut"
+    },
+    {
+      "parameters": {
+        "method": "POST",
+        "url": "={{'http://collector:8081/internal/jobs/steps/real-cut/run?job_id=' + ($json.body.job_id || $json.body.jobId)}}",
+        "sendHeaders": true,
+        "headerParameters": {
+          "parameters": [
+            {
+              "name": "X-Orchestrator-Secret",
+              "value": "storyforge-local-secret"
+            }
+          ]
+        },
+        "options": {
+          "timeout": 3600000
+        }
+      },
+      "id": "realcut-runner",
+      "name": "Run Real Cut Step",
+      "type": "n8n-nodes-base.httpRequest",
+      "typeVersion": 4.2,
+      "position": [
+        520,
+        300
+      ]
+    }
+  ],
+  "connections": {
+    "Real Cut Webhook": {
+      "main": [
+        [
+          {
+            "node": "Run Real Cut Step",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Run Real Cut Step": {
+      "main": [
+        []
+      ]
+    }
+  },
+  "active": false,
+  "settings": {},
+  "pinData": {},
+  "versionId": "storyforge-real-cut-v1"
+}