Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.
Python agent-contextagent-memoryagent-pluginsagentic-ragcontext-databasedsh-pluginknowledge-basememorymemory-managementself-evolvingvector-database
fix: normalize agent logos and guide rendering (#4325)
yufengc3f3fb499ec7
diff --git a/docs/images/agents/en/doubao-work.md b/docs/images/agents/en/doubao-work.mdindex 12d51ff24..f43e8b47e 100644--- a/docs/images/agents/en/doubao-work.md+++ b/docs/images/agents/en/doubao-work.md@@ -1,25 +1,23 @@## Step 1: Install the connector-1. Open Doubao Work, select **Skills · Connectors · Partners** in the sidebar, search for “OpenViking Context,” and select **+**.-+1. Open Doubao Work, select **Skills · Connectors · Partners** in the sidebar, search for “OpenViking Context,” and select <strong>+</strong>.2. Enter the OpenViking USER API key in the authorization dialog:-{{OPENVIKING_API_KEY_BLOCK}}--3. Select **Save and Connect**. Integration is complete when the “Connector installed” message appears and the **+** next to “OpenViking Context” changes to the added state.+ ```text+ {{OPENVIKING_API_KEY}}+ ```+3. Select **Save and Connect**. Integration is complete when the “Connector installed” message appears and the <strong>+</strong> next to “OpenViking Context” changes to the added state.## Step 2: Verify1. Return to a Doubao conversation, select **Connectors** below the input box, and confirm that “OpenViking Context” is available.-2. Select **More Skills**, confirm that “OpenViking Context Database” is available, and ask Doubao to call OpenViking and return relevant content.-## Troubleshootingdiff --git a/docs/images/agents/en/workbuddy.md b/docs/images/agents/en/workbuddy.mdindex dc4988ae5..805eff57e 100644--- a/docs/images/agents/en/workbuddy.md+++ b/docs/images/agents/en/workbuddy.md@@ -1,38 +1,33 @@## Step 1: Configure MCP1. Open WorkBuddy, select **Experts · Skills · Connectors** in the sidebar, and open the **Connectors** tab.-2. Select **Custom Connector** in the upper-right corner to open MCP service management.-3. Select **Configure MCP** to open the MCP configuration editor.-4. Add this configuration:-```json-{- "mcpServers": {- "OpenViking": {- "url": "https://api.vikingdb.cn-beijing.volces.com/openviking/mcp",- "headers": {- "Authorization": "Bearer {{OPENVIKING_API_KEY}}"+ ```json+ {+ "mcpServers": {+ "OpenViking": {+ "url": "https://api.vikingdb.cn-beijing.volces.com/openviking/mcp",+ "headers": {+ "Authorization": "Bearer {{OPENVIKING_API_KEY}}"+ }}- }- }-}-```+ }+ }+ ```5. Select **Save** in the upper-right corner. The configuration is saved when the green success message appears.-6. Return to the MCP list. If WorkBuddy asks you to trust this MCP service on first connection, select **Trust**.-## Step 2: Verify@@ -46,4 +41,4 @@ Return to the MCP list. Confirm that `OpenViking` appears under “My MCP,” is| Issue | Resolution ||---|---|| `OpenViking` does not appear in the MCP list | Check the JSON syntax and save again |-| MCP connection status is abnormal | Refresh the connection; if it still fails, check the JSON configuration, API key, and network connectivity |+| MCP connection status is abnormal | Refresh the connection; if it still fails, check the JSON configuration and network connectivity |diff --git a/docs/images/agents/image/deepseek-harness/logo.png b/docs/images/agents/image/deepseek-harness/logo.pngindex e0875dc6e..a331b221e 100644Binary files a/docs/images/agents/image/deepseek-harness/logo.png and b/docs/images/agents/image/deepseek-harness/logo.png differdiff --git a/docs/images/agents/image/doubao-work/logo.png b/docs/images/agents/image/doubao-work/logo.pngindex f1ab4e8d0..904af3798 100644Binary files a/docs/images/agents/image/doubao-work/logo.png and b/docs/images/agents/image/doubao-work/logo.png differdiff --git a/docs/images/agents/image/workbuddy/logo.png b/docs/images/agents/image/workbuddy/logo.pngindex e02706978..18330882e 100644Binary files a/docs/images/agents/image/workbuddy/logo.png and b/docs/images/agents/image/workbuddy/logo.png differdiff --git a/docs/images/agents/zh/doubao-work.md b/docs/images/agents/zh/doubao-work.mdindex d21ca5d75..410a98d67 100644--- a/docs/images/agents/zh/doubao-work.md+++ b/docs/images/agents/zh/doubao-work.md@@ -1,25 +1,23 @@## 步骤1:安装连接器-1. 打开豆包工作,点击左侧导航栏的 **技能·连接器·伙伴**,搜索“OpenViking Context”,点击右侧的 **+**。-+1. 打开豆包工作,点击左侧导航栏的 **技能·连接器·伙伴**,搜索“OpenViking Context”,点击右侧的 <strong>+</strong>。2. 在“授权配置”窗口中填写 OpenViking USER API Key:-{{OPENVIKING_API_KEY_BLOCK}}--3. 点击 **保存并连接**。页面顶部出现“连接器已安装”提示,且“OpenViking Context”右侧由 **+** 变为已添加状态,即表示接入完成。+ ```text+ {{OPENVIKING_API_KEY}}+ ```+3. 点击 **保存并连接**。页面顶部出现“连接器已安装”提示,且“OpenViking Context”右侧由 <strong>+</strong> 变为已添加状态,即表示接入完成。## 步骤2:验证1. 返回豆包主对话,点击对话框下方的 **连接器**,确认能够找到“OpenViking Context”。-2. 点击对话框下方的 **更多技能**,确认能够找到“OpenViking 上下文数据库”,并让豆包调用 OpenViking 返回相关内容。-## 故障排查diff --git a/docs/images/agents/zh/workbuddy.md b/docs/images/agents/zh/workbuddy.mdindex 5ee26f702..59557c62b 100644--- a/docs/images/agents/zh/workbuddy.md+++ b/docs/images/agents/zh/workbuddy.md@@ -1,38 +1,33 @@## 步骤1:配置 MCP1. 打开 WorkBuddy,点击左侧导航栏的 **专家·技能·连接器**,进入 **连接器** 标签页。-2. 点击右上角的 **自定义连接器**,进入 MCP 服务管理面板。-3. 点击 **配置 MCP**,进入 MCP 配置编辑器。-4. 在配置文件中填入以下内容:-```json-{- "mcpServers": {- "OpenViking": {- "url": "https://api.vikingdb.cn-beijing.volces.com/openviking/mcp",- "headers": {- "Authorization": "Bearer {{OPENVIKING_API_KEY}}"+ ```json+ {+ "mcpServers": {+ "OpenViking": {+ "url": "https://api.vikingdb.cn-beijing.volces.com/openviking/mcp",+ "headers": {+ "Authorization": "Bearer {{OPENVIKING_API_KEY}}"+ }}- }- }-}-```+ }+ }+ ```5. 点击右上角的 **保存**。顶部出现“配置保存成功”的绿色提示后,配置即已保存。-6. 返回 MCP 列表。如果系统提示“首次连接此 MCP 服务需要您的信任确认”,点击 **信任** 完成接入。-## 步骤2:验证@@ -46,4 +41,4 @@| 问题 | 处理 ||---|---|| MCP 列表中未出现 `OpenViking` | 检查 JSON 配置格式并重新保存 |-| MCP 连接状态异常 | 刷新连接;若仍异常,检查 JSON 配置、API Key 和网络连通性 |+| MCP 连接状态异常 | 刷新连接;若仍异常,检查 JSON 配置及网络连通性 |