Changelog
追踪 WaoClaw 的每一次更新。新功能、改进、修复，透明公开。
Static Page Deployment
Subscription users now support automated deployment and sharing of static pages
MCP Service Management
npx-based MCP services now automatically optimize their startup paths, the service list now shows installation/optimization status, and retry is supported after failures
Media References Resolution
@ media references in media generation are now resolved more reliably into actual file paths or URLs, reducing failures caused by incorrect reference images, first frames, or reference videos
Batch Deletion for Subtasks
Batch deletion is now supported for subtasks/sub-agent sessions, and the corresponding gateway sessions will be cleaned up in the background after deletion
Expert Kits
Allowing users to install and use grouped capabilities with one click based on work scenarios
Fork Conversation
Allowing users to copy the current conversation history and start a new exploration branch
Check for Updates
Added to plugin settings page, supporting manual detection and updates for npm and ClawHub plugins
Quick Account Menu
Added quick entries for usage overview, top-up, inviting friends to earn credits, and more
MiniMax M3 Model
Added as a default model and updated the default context length for some custom models
Large Content Preview
Large Markdown content and tool outputs now use a lightweight preview by default and can be expanded on demand, reducing page lag
Video Generation Dialog
Optimized to display consumption information more clearly, making it easier to confirm usage before submission
WeChat Login State Loss
Fixed an issue where the WeChat login state was lost after an overwrite installation on Windows
MCP Runtime
Optimized MCP runtime reuse and process cleanup, reducing background process buildup when using multiple conversations
AI Engine Connection Interruptions
Fixed false reports of AI engine connection interruptions during large-output tool execution, and reduced unnecessary engine reloads during startup
Image Generation
Supporting Doubao-Seedream, Wan 2.7 image, and Minimax image series models
Video Generation
Supporting Doubao-Seedance 2.0, HappyHorse, Kling, and Minimax Hailuo series models
Plugin Market Sync
Plugin market now supports syncing plugins from the OpenClaw extensions directory
Tool Loop Detector
Added a tool loop detector to prevent aborted tool loops from continuing to consume tokens
Gateway sessions.patch Timeout
Fixed the issue where OpenClaw gateway sessions.patch timeout blocked chat.send
GitHub Copilot Token Refresh
Fixed the issue where GitHub Copilot token refresh triggered a gateway restart
Sub-agent Cleanup Blocked
Fixed the issue where the sub-agent cleanup phase was blocked due to a hook failure
Custom Model Parameters
Custom parameters in JSON format can now be configured individually for each model, making it easy to control advanced settings like thinking depth
Thinking Process Display
The model thinking process is presented in an independent module, clearly separated from the final response
Sub-agent Sidebar
Sub-agent conversations are displayed independently in the sidebar, maintaining consistent styling and interaction with the main conversation
Artifacts Preview
The preview browser now features more powerful annotation capabilities, facilitating direct markup and collaboration on artifacts
Multi-instance Bot Management UI
Redesigned, making multi-account management clearer and more intuitive
Qwen 3.6 Plus Coding Plan
Added support for the Qwen 3.6 Plus Coding Plan
Browser Network Access
Added browser network access mode selection and domain blocking configurations
Model Context Window
Added custom settings support for model context window
Multi-tab Artifacts
Added multi-tab support for the Artifacts right-side preview
Pairing Code Support
Added pairing code support for more IM platforms
POPO Plugin Upgrade
Upgraded to version 2.1.8
Optimizations
Optimized Dreaming and scheduled tasks
Xiaomi Mimo Compatibility
Fixed compatibility issues with the Xiaomi Mimo model
MCP Service Naming
Fixed a bug caused by non-ASCII characters in MCP service name
Context Usage Visualization
Supporting manual compression of conversation context
Plugin Management Panel
Supporting online installation, uninstallation, and advanced configuration
Excel & PPTX Preview
Significantly improved, supporting quick opening with local applications
MCP Protocol
Reused native OpenClaw MCP protocol for more stable and reliable external tool connections
Heartbeat Token Consumption
Reduced token consumption for heartbeat messages
Tool Result Images
Images from tool execution results are now correctly displayed in the conversation
File Preview System
Files generated in conversations are uniformly displayed via preview system
IM Channel Working Directory
Fixed an issue where the IM channel working directory was incorrect
MCP Server Auth
Fixed an issue where MCP servers failed authentication due to duplicate headers
MCP stdio Startup
Fixed an issue where MCP stdio-type servers failed to start due to environment variable problems
Dreaming Memory Consolidation
Supporting background memory organization toggle, scheduling strategy, timezone, and model configuration
Memory Settings Page
Refactored into tabbed layout, added Dreaming status and diary content display
Youdao Cloud Notes Skill
Upgraded to v1.0.9
Artifacts Preview
Now supports Mermaid zoom controls, auto-refresh on external file changes, and hides Code tab for binary files
Copy Images to Clipboard
Added support for copying PNG/JPEG images to clipboard
UI Details
Improved code block component styling, model selector placement, and layout visual details
IM Channel Config
Fixed IM channel configuration not taking effect after switching Agent
Preview Anomalies
Fixed Mermaid/PPTX preview anomalies and Ctrl+scroll zoom not working
Chinese Filename in Browser
Fixed files with Chinese-character paths failing to open in browser on Windows
Various Bug Fixes
Fixed icon display, NO_REPLY tag stripping, bulk delete operations, and more
Artifacts Preview System
Supporting real-time preview of HTML/SVG/React/Mermaid, as well as Office and PDF files
Independent Working Directory
Each Agent supports configuring an independent working directory
Cron Expression Support
Added Cron expression type for scheduled tasks, with paginated filtering for execution history
Code Highlighting Engine
Upgraded to CodeMirror 6, supporting search and dark mode
Message Metadata
Assistant messages now display timestamp and model information
Streamed Text Characters
Fixed an issue where characters were occasionally dropped when merging streamed text
Tool Authorization Prompts
Fixed an issue where tool authorization prompts still appeared for stopped conversations
Workspace Decoupling
Decoupled the main Agent workspace from the working directory
Multiple POPO Bots
Added support for multiple POPO bots
Youdao Note Skill
Upgraded
Markdown Table Rendering
Fixed an issue where Markdown tables occasionally failed to render and table contents misaligned
Model Switching in IM
Fixed an issue where switching models in IM tasks did not take effect
Qwen 3.6 Plus Image Input
Fixed an issue where Qwen 3.6 Plus did not support image input
WeCom Configurations
Fixed an issue where WeCom dmPolicy/allowFrom configurations were hardcoded
Windows EPERM Errors
Fixed occasional EPERM permission errors when deleting/installing skills on Windows
ChatGPT OAuth Login
OpenAI model provider supports ChatGPT OAuth login
Coding Plan Support
Xiaomi MiMo / Baidu Qianfan model providers support Coding Plan
WeChat Plugin
Upgraded OpenClaw-WeiXin to 2.1.10
Log Security
Automatically mask sensitive fields in logs
Gateway Restarts
Fixed unexpected Gateway restarts in various scenarios
DeepSeek V4 Custom Providers
Fixed issues when using DeepSeek V4 with custom providers
IM Issues
Fixed incorrect IM history sorting and garbled Chinese filenames in Feishu
Startup & Windows Installer
Increased startup timeout and fixed Windows installer exit code evaluation
Remote Embedding Service
Supporting OpenAI, Gemini, etc. for memory semantic search
DeepSeek V4 Thinking Mode
Fixed compatibility when tool calls are present (reasoning_content)
OpenAI System Proxy
Fixed issue where OpenAI native models become unavailable when system proxy is enabled
DiffView Rendering
Fixed issue where DiffView fails to render when edits array is passed
Skill Discovery Path
Fixed Skill discovery path to align with actual runtime resolution path
CJK Word Segmentation
Fixed CJK word segmentation issue in memory search on Windows
Session Timeout
Fixed old timeout callback incorrectly ending new conversation sessions
New Models
Added support for deepseek-v4-flash, deepseek-v4-pro, and kimi-k2.6 models
Search Experience
Optimized skills and MCP search experience
Multi-instance Bots
Discord and Telegram now support configuring multiple bot instances
LM Studio Provider
Added LM Studio provider to model configuration
Plugin Upgrades
WeCom and POPO plugin upgrades
Draft Preservation
Preserve draft input and attachments when switching sessions
Sharp Native Binding
Fixed sharp native binding issue, added cfmind node_modules to NODE_PATH
MCP Tool Timeout
Fixed stale reply infinite loop when MCP tool times out
MCP Bridge Error
Fixed incorrect request termination caused by MCP bridge listener object error
Image Model Recognition
Fixed some image-capable models unable to recognize image messages
OpenClaw Upgrade
Upgraded OpenClaw to v2026.4.14
WaoClaw Email Channel
Added clawEmail IM channel
DingTalk QR Binding
Added QR code scanning binding method for DingTalk bot configuration
Database Recovery
Automatically attempt to recover from backup during database exceptions
Welcome Guide
Added welcome guide pop-up upon first launch
Log Export
Included OpenClaw gateway logs when exporting log ZIP package
Cold Start Optimization
Significantly shortened cold start time by displaying the window earlier and delaying Skill service initialization
DingTalk Connector
Fixed startup crash by downgrading to stable version 0.8.16
Multiple IM Sync Issues
Fixed multiple IM message synchronization anomalies
OpenClaw Engine Upgrade
Major OpenClaw Engine Upgrade (v2026.3.2 → v2026.4.8): covering core improvements in tool-calling stability, multi-agent collaboration, task scheduling, and more
Gateway Startup Speed
Significantly improved: disabled 64 unused built-in plugins, turned off Bonjour service, enabled esbuild minification
Session Retention
Optimized to 30 days by default, eliminating the need to re-establish context during continuous work
WeCom Multiple Bots
WeCom can now integrate with multiple bots for different business scenarios
POPO QR Login
Added QR code scanning method to log in to POPO
Baidu Qianfan LLM
Added Qianfan integration, allowing direct usage within model settings
MCP Tool Fixes
Resolved MCP tool unavailability in packaged version; MCP config changes now auto-trigger restart
DingTalk Migration
Migrated to official connector for more stable connection
Multi-IM Channel Fixes
Comprehensive fixes for Feishu, QQ, Telegram, Discord, WeCom connection logic
WeChat Channel
Fixed WeChat QR code login and channel registration failures
Scheduled Task Fixes
Fixed cross-timezone task time drift, WeCom notification failures, unsaved prompt logic
MiniMax Auth UI
Completely revamped authentication interface using Radio Cards instead of Tabs
Model Provider Registry
Unified and integrated all model provider metadata into a shared registry
Skill Creation on Windows
Fixed incorrect path prompts causing files to be written to wrong directory
Scheduled Task Dirty State
Fixed issue where dirty state was not reset after saving a scheduled task
Session Keep-Alive
Optimized OpenClaw session keep-alive strategy for long-running tasks
Outdated Provider Config
Fixed issue where outdated Provider configurations remained after deleting all models
Bug Fixes
Various bug fixes and stability improvements