Fix/bot Fix Memory Consolidation Issues and Integrate Vikingbot as OpenViking Optional Dependency (#492)

* fix

* add log

* add log

* update

* Opt memory

* Opt memory

* Opt memory

* Opt memory

* Fix memory consolidation issues: duplicate session write, hook error, and async

- Fix _consolidate_memory hook error by creating temporary Session when given message list
- Remove duplicate session save logic from _consolidate_memory (handled by caller)
- Make normal memory consolidation async like /new command
- Other fixes: streaming response in CLI and OpenAPI channel

* Integrate vikingbot as openviking optional dependency

- Add vikingbot dependencies to pyproject.toml optional-dependencies
- Update install prompts from vikingbot[X] to openviking[bot-X]
- Update README installation instructions
- Configure setuptools to find vikingbot in bot/ directory
- Add vikingbot package data and script entry

* Remove bot/pyproject.toml

vikingbot is now integrated as part of openviking, no need for separate pyproject.toml

* Update vikingbot installation instructions in root READMEs

- Update from 'uv pip install -e bot/' to 'uv pip install -e ".[bot]"'
- Add quotes around openviking[bot] for shell safety

* update

* Use singleton pattern for VikingClient in hooks

Cache VikingClient instances by workspace_id to avoid repeated client creation
- Add _client_cache dictionary for caching
- Add get_cached_client() helper function
- Update both hooks to use cached client

* Use global singleton for VikingClient instead of per-workspace cache

- Change from per-workspace_id cache to true global singleton
- Create client with None as agent_id (works for all workspaces)
- Simplify client management

* uv run ruff format

---------

Co-authored-by: DuTao <dutao.1786@bytedance.com>

Author: chenjw

README.md

@@ -443,10 +443,10 @@ VikingBot is an AI agent framework built on top of OpenViking. Here's how to get
 
 ```bash
 # Option 1: Install VikingBot from PyPI (recommended for most users)
-pip install openviking[bot]
+pip install "openviking[bot]"
 
 # Option 2: Install VikingBot from source (for development)
-uv pip install -e bot/
+uv pip install -e ".[bot]"
 
 # Start OpenViking server with Bot enabled
 openviking-server --with-bot

README_CN.md

@@ -443,10 +443,10 @@ VikingBot 是构建在 OpenViking 之上的 AI 智能体框架。以下是快速
 
 ```bash
 # 选项 1：从 PyPI 安装 VikingBot（推荐大多数用户使用）
-pip install openviking[bot]
+pip install "openviking[bot]"
 
 # 选项 2：从源码安装 VikingBot（用于开发）
-uv pip install -e bot/
+uv pip install -e ".[bot]"
 
 # 启动 OpenViking 服务器（同时启动 Bot）
 openviking-server --with-bot

bot/README.md

@@ -16,7 +16,7 @@ Vikingbot is deeply integrated with OpenViking, providing powerful knowledge man
 
 **Option 1: Install from PyPI (Simplest)**
 ```bash
-pip install vikingbot
+pip install "openviking[bot]"
 ```
 
 **Option 2: Install from source (for development)**
@@ -37,7 +37,7 @@ powershell -c "irm https://astral.sh/uv/install.ps1 | iex"
 
 ```bash
 git clone https://github.com/volcengine/OpenViking
-cd OpenViking/bot
+cd OpenViking
 
 # Create a virtual environment using Python 3.11 or higher
 uv venv --python 3.11
@@ -47,10 +47,10 @@ source .venv/bin/activate  # macOS/Linux
 # .venv\Scripts\activate   # Windows
 
 # Install dependencies (minimal)
-uv pip install -e .
+uv pip install -e ".[bot]"
 
 # Or install with optional features
-uv pip install -e ".[langfuse,telegram,console]"
+uv pip install -e ".[bot,bot-langfuse,bot-telegram]"
 ```
 
 ### Optional Dependencies
@@ -59,25 +59,25 @@ Install only the features you need:
 
 | Feature Group | Install Command | Description |
 |---------------|-----------------|-------------|
-| **Full** | `uv pip install -e ".[full]"` | All features included |
-| **Langfuse** | `uv pip install -e ".[langfuse]"` | LLM observability and tracing |
-| **FUSE** | `uv pip install -e ".[fuse]"` | OpenViking filesystem mount |
-| **Sandbox** | `uv pip install -e ".[sandbox]"` | Code execution sandbox |
-| **OpenCode** | `uv pip install -e ".[opencode]"` | OpenCode AI integration |
+| **Full** | `uv pip install -e ".[bot-full]"` | All features included |
+| **Langfuse** | `uv pip install -e ".[bot-langfuse]"` | LLM observability and tracing |
+| **FUSE** | `uv pip install -e ".[bot-fuse]"` | OpenViking filesystem mount |
+| **Sandbox** | `uv pip install -e ".[bot-sandbox]"` | Code execution sandbox |
+| **OpenCode** | `uv pip install -e ".[bot-opencode]"` | OpenCode AI integration |
 
 #### Channels (chat apps)
 
 | Channel | Install Command |
 |---------|-----------------|
-| **Telegram** | `uv pip install -e ".[telegram]"` |
-| **Feishu/Lark** | `uv pip install -e ".[feishu]"` |
-| **DingTalk** | `uv pip install -e ".[dingtalk]"` |
-| **Slack** | `uv pip install -e ".[slack]"` |
-| **QQ** | `uv pip install -e ".[qq]"` |
+| **Telegram** | `uv pip install -e ".[bot-telegram]"` |
+| **Feishu/Lark** | `uv pip install -e ".[bot-feishu]"` |
+| **DingTalk** | `uv pip install -e ".[bot-dingtalk]"` |
+| **Slack** | `uv pip install -e ".[bot-slack]"` |
+| **QQ** | `uv pip install -e ".[bot-qq]"` |
 
 Multiple features can be combined:
 ```bash
-uv pip install -e ".[langfuse,telegram,console]"
+uv pip install -e ".[bot,bot-langfuse,bot-telegram]"
 ```
 
 ## 🚀 Quick Start
@@ -434,7 +434,7 @@ For Langfuse Cloud, use `https://cloud.langfuse.com` as the `base_url`.
 
 **Install Langfuse support:**
 ```bash
-uv pip install -e ".[langfuse]"
+uv pip install -e ".[bot-langfuse]"
 ```
 
 **Restart vikingbot:**

bot/README_CN.md

@@ -17,7 +17,7 @@ Vikingbot 深度集成 OpenViking，提供强大的知识管理和记忆检索
 
 **选项 1：从 PyPI 安装（最简单）**
 ```bash
-pip install vikingbot
+pip install "openviking[bot]"
 ```
 
 **选项 2：从源码安装（用于开发）**
@@ -38,7 +38,7 @@ powershell -c "irm https://astral.sh/uv/install.ps1 | iex"
 
 ```bash
 git clone https://github.com/volcengine/OpenViking
-cd OpenViking/bot
+cd OpenViking
 
 # 创建 Python 3.11 或更高版本 虚拟环境
 uv venv --python 3.11
@@ -48,10 +48,10 @@ source .venv/bin/activate  # macOS/Linux
 # .venv\Scripts\activate   # Windows
 
 # 安装依赖（最小化）
-uv pip install -e .
+uv pip install -e ".[bot]"
 
 # 或安装包含可选功能
-uv pip install -e ".[langfuse,telegram,console]"
+uv pip install -e ".[bot,bot-langfuse,bot-telegram]"
 ```
 
 ### 可选依赖
@@ -60,25 +60,25 @@ uv pip install -e ".[langfuse,telegram,console]"
 
 | 功能组 | 安装命令 | 描述 |
 |---------------|-----------------|-------------|
-| **完整版** | `uv pip install -e ".[full]"` | 包含所有功能 |
-| **Langfuse** | `uv pip install -e ".[langfuse]"` | LLM 可观测性和追踪 |
-| **FUSE** | `uv pip install -e ".[fuse]"` | OpenViking 文件系统挂载 |
-| **沙箱** | `uv pip install -e ".[sandbox]"` | 代码执行沙箱 |
-| **OpenCode** | `uv pip install -e ".[opencode]"` | OpenCode AI 集成 |
+| **完整版** | `uv pip install -e ".[bot-full]"` | 包含所有功能 |
+| **Langfuse** | `uv pip install -e ".[bot-langfuse]"` | LLM 可观测性和追踪 |
+| **FUSE** | `uv pip install -e ".[bot-fuse]"` | OpenViking 文件系统挂载 |
+| **沙箱** | `uv pip install -e ".[bot-sandbox]"` | 代码执行沙箱 |
+| **OpenCode** | `uv pip install -e ".[bot-opencode]"` | OpenCode AI 集成 |
 
 #### 聊天渠道
 
 | 渠道 | 安装命令 |
 |---------|-----------------|
-| **Telegram** | `uv pip install -e ".[telegram]"` |
-| **飞书/Lark** | `uv pip install -e ".[feishu]"` |
-| **钉钉** | `uv pip install -e ".[dingtalk]"` |
-| **Slack** | `uv pip install -e ".[slack]"` |
-| **QQ** | `uv pip install -e ".[qq]"` |
+| **Telegram** | `uv pip install -e ".[bot-telegram]"` |
+| **飞书/Lark** | `uv pip install -e ".[bot-feishu]"` |
+| **钉钉** | `uv pip install -e ".[bot-dingtalk]"` |
+| **Slack** | `uv pip install -e ".[bot-slack]"` |
+| **QQ** | `uv pip install -e ".[bot-qq]"` |
 
 可以组合多个功能：
 ```bash
-uv pip install -e ".[langfuse,telegram,console]"
+uv pip install -e ".[bot,bot-langfuse,bot-telegram]"
 ```
 
 ## 🚀 快速开始
@@ -430,7 +430,7 @@ cd deploy/docker
 
 **安装 Langfuse 支持：**
 ```bash
-uv pip install -e ".[langfuse]"
+uv pip install -e ".[bot-langfuse]"
 ```
 
 **重启 vikingbot：**

bot/pyproject.toml

@@ -7,7 +7,7 @@ set -e
 
 # Default values
 PORT="1933"
-BOT_URL="http://localhost:18791"
+BOT_URL="http://localhost:18790"
 
 # Parse arguments
 while [[ $# -gt 0 ]]; do
@@ -31,7 +31,7 @@ done
 # Parse Bot URL to extract port
 BOT_PORT=$(echo "$BOT_URL" | sed -n 's/.*:\([0-9]*\).*/\1/p')
 if [ -z "$BOT_PORT" ]; then
-    BOT_PORT="18791"
+    BOT_PORT="18790"
 fi
 
 echo "=========================================="

bot/scripts/restart_openviking_server.sh

@@ -0,0 +1,47 @@
+import openviking as ov
+
+# Initialize OpenViking client with data directory
+client = ov.SyncOpenViking(path="./data")
+
+try:
+    # Initialize the client
+    client.initialize()
+
+    # Add resource (supports URL, file, or directory)
+    add_result = client.add_resource(
+        path="/Users/bytedance/Downloads/exp/experience_data_mini.json",
+        resource_type="json",  # 明确指定类型
+        tags=["large_data", "agent_context", "structured"],
+    )
+    root_uri = add_result["root_uri"]
+
+    # Explore the resource tree structure
+    ls_result = client.ls(root_uri)
+    print(f"Directory structure:\n{ls_result}\n")
+
+    # Use glob to find markdown files
+    glob_result = client.glob(pattern="**/*.md", uri=root_uri)
+    if glob_result["matches"]:
+        content = client.read(glob_result["matches"][0])
+        print(f"Content preview: {content[:200]}...\n")
+
+    # Wait for semantic processing to complete
+    print("Wait for semantic processing...")
+    client.wait_processed()
+
+    # Get abstract and overview of the resource
+    abstract = client.abstract(root_uri)
+    overview = client.overview(root_uri)
+    print(f"Abstract:\n{abstract}\n\nOverview:\n{overview}\n")
+
+    # Perform semantic search
+    results = client.find("what is openviking", target_uri=root_uri)
+    print("Search results:")
+    for r in results.resources:
+        print(f"  {r.uri} (score: {r.score:.4f})")
+
+    # Close the client
+    client.close()
+
+except Exception as e:
+    print(f"Error: {e}")