Skip to main content

Changelog


[1.13.0] - 2025/05/16


Core Feature Upgrades

Agent UI Cross-Platform Voice Capability Enhancement

  • Supports switching between 4 scenario-based voices (chat/reading/consultation/general)
  • Adds playback support for silent mode
  • Mini Program & Low-code components support voice speed adjustment
  • Optimized dual-title display logic (hiding redundant titles in the content area)
  • Resolved jitter issues in text-image mixed layouts

Agent UI Web Component Capability Expansion

  • Adds voice playback function
  • Supports three types of tool cards including map/weather
  • Fixed multiple interaction issues

🌐 Expand Open Capabilities

MCP Server Template Expansion

  • Tencent Hunyuan 3D MCP: Supports generating 3D models from text/images (MAX/FBX/GLB formats)
  • MarkItDown Service MCP: Implements conversion from any file to Markdown format
  • CRIC Property AI MCP: Integrates dedicated intelligent assistant capabilities for the property management industry

Functional Agent Expansion

  • LKE Functional Agent Template: Supports integration with large model knowledge engines

🚧 New Capabilities Preview

  • Agent interface allows users to switch models independently.
  • Optimization of knowledge base retrieval and compatibility with multiple file formats
  • Integration of Official Account / Service Account / WeChat Customer Service with Functional Agent capabilities
  • Agent supports integration with more third-party Agent development platforms
  • Supports one-click conversion of OpenAPI to MCP
  • MCP that provides integration with third-party knowledge base retrieval and Agent platforms

Features will be rolled out in batches. Please stay tuned for the specific schedule in version update notifications.

[1.12.0] - 2025/04/29


Core Feature Upgrades

Agent supports voice input/output capabilities

Agent natively integrates voice input/output capabilities, enabling developers to rapidly build multimodal interaction scenarios.

Agent supports multi-session capability

Agent supports handling multiple independent sessions simultaneously within the same instance, with each session maintaining an independent history.

Agent UI Mini Program Component supports tool cards

Agent UI mini-program source code supports ToolCard tool cards, enabling flexible customization of MCP display effects. Supports custom business logic such as product recommendations or map navigation.


⚙️ Developer Experience Improvements

File Upload Component Security Upgrade

Due to the security policy upgrade in the WeChat Mini Program basic library, the file upload feature parsing process may be affected. Please upgrade the Agent UI component.

  • Mini Program source code component, please update to the latest version.
  • For WeDa low-code components, create a new page and re-drag to update the agent-ui block, then publish the application after completion.

🚧 New Capabilities Preview

  • Agent interface allows users to switch models independently.
  • Optimization of knowledge base retrieval and compatibility with multiple file formats
  • Integration of Official Account / Service Account / WeChat Customer Service with Functional Agent capabilities
  • Agent supports integration with more third-party Agent development platforms
  • Supports one-click conversion of OpenAPI to MCP
  • MCP that provides integration with third-party knowledge base retrieval and Agent platforms

Features will be rolled out in batches. Please stay tuned for the specific schedule in version update notifications.

[1.11.0] - 2025/04/17

Core Feature Upgrades

  • Supports zero-barrier cloud hosting for MCP Server

    • Comprehensive support for MCP Servers written in mainstream languages such as Java/Python/GoLang/PHP, without invasive framework modifications

    Local MCP Servers can seamlessly migrate to CloudBase hosting, automatically gaining access to secure and reliable remote MCP services.

  • The CloudBase MCP Marketplace provides a standardized listing process

🌐 Expand Open Capabilities

  • CloudBase MCP Marketplace is now officially launched

    CloudBase provides a one-stop development and hosting platform for MCP Servers, supporting rapid publishing and deployment of MCP Servers.

    Added 30+ standardized service modules, covering mainstream development scenarios:

    DomainRepresentative Service Modules
    AI AugmentationCloudBase AI Capabilities / Memory Module / Meta Plugin / Puppeteer Automation
    Data ServicesMongoDB/MySQL/Postgres/Redis Database Access / COS File and Data Processing
    Developer ToolsGithub/Gitee/Gitlab Code Management / Figma Design Collaboration / Apifox API Debugging
    Maps and Lifestyle ServicesTencent Location Service / Amap / Caiyun Weather / ALAPI Short URL and Data Hotlist
    Productivity ToolsNotion Knowledge Base / QuickChart Visualization / Mermaid Chart Rendering / Zapier Universal Workflow
    Cloud-Native InfrastructureCloudBase Environment Management / Lighthouse Cloud Servers / CNB Cloud-Native Build / Trends Hub Hotspot Analysis

🚧 New Capabilities Preview

  • Agent supports session functionality
  • Agent voice input and output capabilities are now available
  • Agent interface allows users to switch models independently.
  • Optimization of knowledge base retrieval and compatibility with multiple file formats
  • Integration of Official Account / Service Account / WeChat Customer Service with Functional Agent capabilities
  • Agent supports integration with more third-party Agent development platforms
  • Supports one-click conversion of OpenAPI to MCP
  • MCP that provides integration with third-party knowledge base retrieval and Agent platforms

Features will be rolled out in batches. Please stay tuned for the specific schedule in version update notifications.

[1.10.0] - 2025/03/28

Core Feature Upgrades: Full Support for MCP

  • Agent supports the MCP protocol
    Agent supports adding MCP Server. AI can automatically select tools based on task requirements and complete complex operations by invoking tools multiple times.

  • CloudBase Supports Development and Hosting of MCP Services
    CloudBase provides a one-stop service from development to deployment, enabling you to host and build your own MCP services on the platform. Provides 6 built-in MCP Server templates

    • CloudBase AI capabilities: Provide atomic AI capabilities via the MCP protocol, such as knowledge base retrieval, database query, large model invocation, web search, and image parsing
    • Yuanqi Plugin: Provides built-in access to 19 official Tencent Yuanqi plugins, supporting capabilities such as image generation, audio generation, and Sogou search. - Yuanqi Agent: Converts Tencent Yuanqi Agent into MCP plugins for use, enabling integration with personal WeChat Official Account articles (Yuanqi Agent has a built-in capacity of 100 million tokens).
    • Memory Module: A module based on CloudBase database + knowledge graph that provides memory for large models and supports storing user-level information
    • Puppeteer: Can control the Chrome browser to perform operations
    • Blank Template: Developers can quickly implement their own MCP Server, enabling Agents to connect to proprietary data and functions
  • Agent UI Components Upgrade

    • Mini Program Agent supports shared CloudBase environment invocation
    • Added status display for the MCP tool invocation process

🚧 New Capabilities Preview

  • Quick import of open-source MCP Servers
  • Agent supports session functionality
  • Agent voice input and output capabilities are now available
  • Agent interface allows users to switch models independently.
  • Optimization of knowledge base retrieval and compatibility with multiple file formats
  • Integration of Official Account / Service Account / WeChat Customer Service with Functional Agent capabilities

Please stay tuned for further updates regarding the specific launch date


[1.9.0] - 2025/03/21

Core Feature Upgrades

  • Agent supports image upload, parsing, and Q&A
    Added support for uploading images in conversations for parsing and generating AI responses, enhancing multimedia interaction capabilities.
  • Agent supports multiple knowledge base selection
    Supports selecting up to 3 knowledge bases during conversations for comprehensive search, enhancing response accuracy and information coverage.

⚙️ System Improvements

  • Knowledge Base Upload and Management Feature Upgrade
    Supports drag-and-drop file upload for more convenient operation. Supports batch upload of up to 200 files. Added batch file deletion feature to improve management efficiency.

  • Error Handling and Prompt Optimization
    Enhanced error message prompts provide clear guidance for situations such as Token expiration and service overlimit. Identify the WeCom environment and display compatibility prompts.

  • Dialogue Processing Logic Enhancement
    Optimize the dialogue logic for WeChat Official Accounts Platform and fix the issue of potentially repeating responses to the same question. Refactor the file content parsing interface to resolve timeout issues during complex file parsing.


🌐 Expand Open Capabilities


🚧 New Capabilities Preview

  • Agent autonomously invokes external tools to implement complex business logic.
  • Agent supports session functionality
  • Agent voice input and output capabilities are now available
  • Agent interface allows users to switch models independently.
  • Optimization of knowledge base retrieval and compatibility with multiple file formats
  • Integration of Official Account / Service Account / WeChat Customer Service with Functional Agent capabilities

Features will be rolled out in batches. Please stay tuned for the specific schedule in version update notifications.


[1.8.0] - 2025/03/14

Core Feature Upgrades

  • Agent supports integration with CloudBase data models
    Added support for integration with CloudBase data models, enabling natural language queries on structured data and generation of AI responses.

  • Agent UI feature optimization, now officially open-sourced


⚙️ System Improvements

  • Integrated CloudBase Extension Capabilities for WeChat DevTools IDE Developers can view AI+ usage statistics and develop Agents via CloudBase - Extensions in WeChat IDE.

  • Large Model Usage Statistics Feature Launched
    You can view detailed token usage via Development Platform-Package Usage-Resource Dashboard.

  • Agent File Upload Control Switch
    Since uploading files during conversations incurs additional large model token consumption, a file upload control switch has been added to the Agent. This feature is only available for Entry Package and above. It will be extended to Personal Package after the large model resource package is launched.

  • Dialogue Processing Logic Optimization

    • Automatically trims leading and trailing whitespace from dialogue content in WeChat
    • The query range for cloud history has been optimized to the last 1 day, enhancing the relevance of dialogue context.

🌐 Expand Open Capabilities


🚧 New Capabilities Preview

  • Agent autonomously invokes external tools to implement complex business logic.
  • Agent supports image upload in the conversation interface
  • Agent supports session functionality
  • Agent voice input and output capabilities are now available
  • Agent interface allows users to switch models independently.
  • Agent supports integration with multiple knowledge bases.
  • Optimization of knowledge base retrieval and compatibility with multiple file formats
  • Integration of Official Account / Service Account / WeChat Customer Service with Functional Agent capabilities

Features will be rolled out in batches. Please stay tuned for the specific schedule in version update notifications.


[1.7.0] - 2025/03/07


Core Feature Upgrades

  • Agent UI document context supports file upload
    Added support for users to upload documents as conversation context in dialogues (this release covers Mini Program source code components and visually built Web applications, with adaptation for visually built Mini Programs under development), expanding multimodal interaction scenarios.

  • Cloud Development Environment Resource Sharing
    Supports invoking AI and related resources by sharing cloud development environments under the same account entity.

  • Mini Program Agent UI Component Experience Optimization Supports text selection and copying, enables dynamic height adjustment of input fields based on text content, and exposes more configurable component properties.


🌐 Expand Open Capabilities

  • Functional Agent Provides Standard Interface Documentation The functional Agent framework provides the IBot interface. By implementing a class that satisfies the IBot interface, developers can deploy corresponding Agents. For details, refer to the documentation .

  • Open Invocation of Agent API
    Developers can invoke Agent interfaces via API Keys to enable automated integration with third-party services.

  • Custom UI Output Solution
    Added the capability for Agents to return custom UI structures. Developers can quickly implement personalized interaction designs via the documentation guide.


⚙️ System Improvements

  • Agent Conversation Effectiveness Optimization
    Improved the accuracy of web search results, optimized the weight allocation strategy for configuration information, and enhanced context understanding and answer relevance.

  • Error Messages and Process Optimization
    Optimize the experience during the Agent setup process


🚧 New Capabilities Preview

  • Integrate with CloudBase database as a structured data source, supporting direct reference in conversations.
  • Agent supports image upload in the conversation interface
  • Agent supports session functionality
  • Agent voice input and output capabilities are now available
  • Agent interface allows users to switch models independently.
  • Agent supports integration with multiple knowledge bases.
  • Optimization of knowledge base retrieval and compatibility with multiple file formats
  • Integration of Official Account / Service Account / WeChat Customer Service with Functional Agent capabilities
  • Supports viewing large model invocation usage statistics
  • Provides out-of-the-box Functional Agent templates and solutions

Features will be rolled out in batches. Please stay tuned for the specific schedule in version update notifications.

[1.6.0] - 2025/02/27


Core Feature Upgrades

  • Functional Agent Launch
    Adds Functional Agent for complex logic development, compatible with open-source Agent frameworks, supports integration with external APIs and diverse data sources, and enhances multi-scenario scalability.
  • Hunyuan Turbo S Fast-Thinking Model Launch
    Introduces the Hunyuan Turbo S fast-thinking model, achieving second-level responses (response speed increased by 100%, first-character latency reduced by 44%), complementing existing slow-thinking models like DeepSeek-R1 in capability.

🎨 Interaction Experience Optimization

  • Agent UI Conversation Interface Refresh
    • Refactored Agent UI, optimizing animation feedback for the "thinking" state and online interactions
    • Chat History: Agent UI adds a historical conversation display feature that supports retrospective queries
    • Internet Search Toggle: C-end users can independently disable the internet search functionality
    • Bug Fixes: Resolved the rendering anomaly of Markdown bold text during line breaks

🌐 Expand Open Capabilities

  • Knowledge Base Management API
    Developers can automate the management of knowledge base documents through APIs, supporting functions such as creation, deletion, and batch operations. View Documentation.

⚙️ System Improvements

  • Answer Logic Optimization
    Fixed the issue of repeating user questions in Agent responses and misuse of Markdown formatting, enhancing output accuracy.

🚧 New Capabilities Preview

  • Supports uploading documents and images in the conversation interface for dialogue
  • Support voice input and output capabilities
  • Agent interface allows users to switch models
  • Integrate with CloudBase database as a data source, enabling direct reference to structured data in conversations.
  • Knowledge Base supports automatic synchronization from WeChat Official Account
  • Optimize knowledge base retrieval effectiveness
  • Knowledge Base supports more types of file formats
  • Support invoking the Agent's API interface via API key
  • Support integration of Official Account / Service Account / WeChat Customer Service with Functional Agent

Features will be rolled out in batches. Please stay tuned for the specific schedule in version update notifications.

[1.5.0] - 2025/02/21

Core Feature Upgrades

  • Web Search Capability
    Agent now supports large models like DeepSeek to access real-time internet data retrieval.
  • Dynamic Model Management
    Added model version control feature, supporting specification of different versions of large models such as DeepSeek-V3.

Performance Improvements

  • Document Processing Capability Enhancement
    The file size of the knowledge base has been expanded to 100MB (with Markdown files limited to 10MB).

🌐 Open APIs

  • Comprehensive API Support
    Provide HTTP REST API and OpenAI SDK-compatible interface calling capabilities.

[1.4.0] - 2025/02/21

🧠 Model Ecosystem

  • New Model Integration
    Added official support for DeepSeek R1 and DeepSeek-V3 large models.

🚀 App Development

  • Rapid Deployment Tool
    Provides end-to-end AI application publishing functionality from dialog flow construction to deployment.
  • Interaction Experience Enhancement
    The console has added a real-time preview module for large model conversations.

📱 Cross-Platform Support

  • Mini Program Integration Solution
    Launched the Agent UI component library and industry template suite.

[Historical Versions]

1.3.0

  • Architecture Upgrade
    WeChat Mini Program Basic Library integrates native AI SDK

1.2.0

  • Mini Program Release Compliance Support
    Provides end-to-end guidance for Mini Program algorithm filing.

1.1.0

  • Development Capability Expansion
    • SDK added support for toolchain invocation
    • Seamless AI Integration for Cloud Functions

1.0.0 - Initial Release

  • Basic Capability Building
    • Hunyuan large model API integration
    • Visual Agent Building Framework
    • Multi-platform publishing capability within the WeChat ecosystem (Mini Program/Official Account/WeCom/Customer Service System)
    • RAG Knowledge Base Engine (supports document-based Q&A scenarios)