Change Log

v0.1.37 (July 7, 2026)

https://github.com/import-ai/omnibox/releases/tag/v0.1.37

Web

1. Added context usage display on conversation pages, including used percentage, remaining percentage, and token usage
2. Improved OmniBox AI chat history display by collapsing intermediate steps by default and adding left-side navigation for quickly revisiting conversations
3. Added support for showing each message's creation time and answer time during conversations, and generation duration for merged messages
4. Added built-in default questions to the chat input when there is no visible chat history
5. Improved the Agent 2.1 approval experience, supporting three approval modes: manual, auto-approve, and auto-reject
6. Fixed some known issues

OmniBox AI

1. Agent 2.1 now supports viewing resource content line counts and child resource counts
2. Agent 2.1 added OmniBox product Q&A capabilities
3. Agent 2.1 added long-context capabilities, so long conversations no longer error directly when approaching the model context limit
4. Improved Agent 2.1 memory saving and fixed an issue where Let OmniBox remember did not work in some scenarios

Sharing & Clipping

1. Fixed abnormal saving when sharing content from the Twitter app
2. Fixed an issue where sharing Zhihu content to OmniBox could fail without an error message when the login state was abnormal

Mobile APP

1. Added text descriptions for Web Search and Deep Thinking in the Mobile APP OmniBox AI chat box, aligning with the web experience
2. Added a confirmation dialog when tapping WeChat login in a mobile browser, then redirects to the mini program after confirmation
3. Aligned resource tool call parameters and resource link display in answers with the web experience
4. Resource links in answers now support jumping to the corresponding detail page, and Smart Folders support jumping to the Smart Folder detail page
5. Improved approval mode button display, keeping it on one line in English and truncating overflow text

WeChat/QQ and Other Assistants

1. Fixed an issue where WeChat and QQ message processing could be interrupted when API key permissions included Tag / Search

v0.1.36 (June 30, 2026)

https://github.com/import-ai/omnibox/releases/tag/v0.1.36

Web

1. Improved chapter and title parsing for Douyin videos
2. Fixed an issue where videos saved from the browser's Douyin recommendation page could not link back to the original video
3. Improved rule quota prompt copy in Smart Folders and Search Filters
4. Added a Back to bottom button to conversation pages
5. Fixed an issue where images could not be saved when clipping Xiaohongshu image-and-text content
6. Fixed some known issues

Mobile APP

1. Added support for binding WeChat Assistant in the app
2. Improved rule quota prompt copy in Smart Folders and Search Filters
3. Added a Back to bottom button to conversation pages
4. Fixed some known issues

v0.1.35 (June 23, 2026)

https://github.com/import-ai/omnibox/releases/tag/v0.1.35

Web

1. Fixed an issue where lowering your own permission could return a 404 when a namespace had only one owner
2. Added remark support for API keys, and improved the interaction details for creating, editing, and deleting them
3. Improved the Settings page, Move To dialog, member search dialog, Notification Center refresh experience, and other UI details
4. Fixed issues with X post clipping icons, resource detail images in PNG exports, automatic linking for VFS paths followed by Chinese punctuation, and more
5. Fixed an issue where Agent authorization / approval submit buttons could be clicked repeatedly
6. Fixed some known issues

Mobile APP (v0.1.34)

1. Fixed an issue where Agent authorization / approval could be submitted repeatedly
2. Fixed an issue where VFS paths followed by Chinese punctuation could not be linked automatically
3. Fixed abnormal resource detail images in PNG exports
4. Added clipping icon adaptation for X posts

v0.1.34 (June 16, 2026)

https://github.com/import-ai/omnibox/releases/tag/v0.1.34

Web

1. Added a Locate button to the toolbar, allowing you to quickly locate the currently opened resource in the left resource list
2. Added a Refresh button to the toolbar for quickly refreshing the left resource list
3. Improved the Notification Center refresh icon and loading state
4. Adjusted the shared search box component, supporting one-click clearing of search content
5. Improved the font and color of buttons on the 404 page
6. Improved the empty state display for root directories in the resource list
7. Added fixed expand and collapse arrows to the root directories: Private, Team, System, and Share
8. Moved the Smart Folder entry into the Private and Team menus, and made namespace tier display more consistent
9. Aligned the web Smart Folder experience on small screens with the Mobile APP
10. Improved Open API and SKILL.md
    • Docs: https://www.omnibox.pro/docs/open-api
11. Fixed some known issues

OmniBox AI

1. Fixed an issue where Smart Folders could be used as resource parents
2. Improved the Agent 2.1 trial usage prompt: when trial uses run out, chats automatically switch to Agent 1.1

Parsing

1. Added support for parsing Ximalaya podcasts
2. Split the file parsing task flow to make scheduling clearer for different file types
3. Improved the async scheduling flow

Mobile APP (v0.1.33)

1. Supports exporting files as PNG long images
2. Improved Mobile APP OmniBox AI list refresh and display logic after creating, deleting, or moving resources
3. Fixed some known issues

v0.1.33 (June 9, 2026)

https://github.com/import-ai/omnibox/releases/tag/v0.1.33

OmniBox AI

1. Agent 2.1 now supports loading Skills from .omnibox/skills; executing Python and similar capabilities is not supported yet
2. Improved Agent 2.1 trial usage copy; when trial uses in a Basic namespace run out, new chats switch to Agent 1.1

Web

1. Improved Notification Center status filters, keeping only All and Unread
2. Added Search Filters, supporting resource filtering by tags and other conditions
3. Added Updated time as a Smart Folder filter condition
4. Improved filter condition display for Smart Folders and Search
5. Improved prompt copy when moving resources
6. Added long image export, supporting PNG export
7. Added Tag Permissions and Search Permissions to API key permissions
8. Open API now supports operating resources, tags, search, and more
9. Added OmniBox Skill, supporting resource operations, web collection, and more in products such as Claude Code and Codex
10. Fixed an issue where clicking Regenerate twice in a row in web OmniBox AI conversations could generate duplicate answers with messy formatting
11. Fixed some known issues

File Parsing

1. Improved parsing quality in edge cases

Mobile APP (0.1.32)

1. Improved Notification Center status filters, keeping only All and Unread
2. Added search filters, supporting resource filtering by tags and other conditions
3. Added Updated time as a Smart Folder filter condition
4. Improved filter condition display for Smart Folders and Search
5. Removed the WeChat login entry in the Zhuoyitong environment
6. Fixed extra top spacing on some Android devices
7. Fixed an issue where resource detail pages could appear blank on vivo phones and earlier iOS versions
8. Improved the Mobile APP OmniBox AI chat experience
9. Fixed some known issues

v0.1.32 (June 2, 2026)

https://github.com/import-ai/omnibox/releases/tag/v0.1.32

Web

1. Added Batch Management
   • You can enter Multi-select mode from the left toolbar and select multiple files, folders, or smart folders at once
   • You can use Create folder with selected resources and automatically move those resources into the newly created folder
   • You can batch Move resources, or drag selected resources directly to the target location
   • You can Add selected resources to chat at once, making it easier to ask questions, summarize, or write based on multiple resources
   • You can batch Move to trash; moving a smart folder to trash does not affect the original resources shown inside it
2. Fixed some known issues

v0.1.31 (May 27, 2026)

https://github.com/import-ai/omnibox/releases/tag/v0.1.31

Web

1. Added Smart Folders
2. Improved the Notification Center refresh experience
   • Manual refresh now resets the 10-second auto-refresh timer
   • When there is no new content, refreshing no longer causes list text or the scrollbar position to jitter
3. Improved synchronization between the resource list and resource detail page
   • After actions such as creating a copy, renaming, or undoing deletion, the left resource list and right detail page stay in sync
   • Fixed incorrect resource list positioning in some cases
4. Improved action feedback for read-only permissions in team namespaces
   • Prompts are clearer when moving, uploading, or performing similar actions in a team namespace where you only have view permission
   • Removed unnecessary technical details such as 403 / 409 from some error messages
5. Improved some dark mode display issues
   • Fixed abnormal content title colors in the search dialog
   • Adjusted some underline, disabled state, and button styles
6. Improved namespace name validation prompts on Settings > General
   • When the input does not meet requirements, the prompt is shown below the input field
   • Validation messages are now "Namespace name must be at least 2 characters" and "Namespace name must be at most 64 characters"
7. Improved internationalization on Settings > API Keys
8. Fixed some known issues

OmniBox AI

1. Improved state persistence for OmniBox AI conversations on the web
   • Opening a historical conversation restores its search state
   • Tool selection state, such as web search and deep thinking, is preserved within the same conversation and after page refresh
   • Starting a new conversation clears the tool selection state automatically
2. Fixed an issue where the status stayed pending after a tool call failed or was rejected
3. Fixed an issue in Mobile APP OmniBox AI where, after tapping a citation to open the source file, the title kept showing "Loading"

File Parsing

1. Improved the prompt shown when audio parsing fails
   • The original task error error is no longer shown directly
   • When no valid speech segment is detected, a friendlier prompt is shown: No valid speech content was detected. Please upload a longer, clearer audio file with human speech.
2. Improved the prompt shown when parsing fails because a file is too long
   • The technical message Content too long is no longer shown
   • Users are prompted that The current file content (xxxxx characters) exceeds the system processing limit (32768 characters). Please try splitting the document and uploading it again.
3. Fixed an issue where background tasks could still show as Completed after parsing failed because a file was too long
4. Improved tasks database implementation

Mobile APP (v0.1.30)

1. The Android Pricing page now supports promo codes
2. Fixed an issue where duplicated text input on mobile could fail to save
3. Mobile OmniBox AI is now implemented natively
4. Recording now supports continuing in the background, so switching apps or locking the screen will not automatically interrupt recording
5. Fixed an occasional issue where, during consecutive recordings, a short recording could be parsed as content from the previous longer recording
6. Improved the source-tracing return experience in Mobile APP OmniBox AI: after tapping a source, you can return to the original Q&A page
7. Fixed extra top spacing on file detail and conversation pages on Xiaomi phones
8. Added Smart Folders
9. Fixed some known issues

v0.1.30 (May 18, 2026)

https://github.com/import-ai/omnibox/releases/tag/v0.1.30

Web

1. Improved the search results page experience
   • Supports Command (Ctrl on Windows) + left-click to open a resource in a new window
   • Closing the search dialog preserves the search box content and results, so they are restored when reopened
   • Right-clicking a resource in search results opens the browser's native context menu
   • Press Esc to exit search
2. Added the current namespace version display in the top-left corner of the homepage
3. Fixed an issue where failed video, audio, or PDF parsing still consumed quota
4. Refactored the web sidebar (resource list)
5. Fixed some trash permission issues
6. Fixed some known issues

OmniBox AI

1. Agent 2.1 added memory support. See docs: https://www.omnibox.pro/docs/omnibox-ai/agent-2-1#agent-memory
2. Fixed an issue where Agent 2.1 answers could not cite knowledge base content

Web Clipping

1. Fixed an issue where clipping X (Twitter) posts could report an error
   • Note: parsing videos in posts is not supported
2. Fixed some known issues

User Docs

1. Added "Notification Center" docs: https://www.omnibox.pro/docs/notification-center
2. Added "Agent 2.1" docs: https://www.omnibox.pro/docs/omnibox-ai/agent-2-1

v0.1.29 (April 30, 2026)

https://github.com/import-ai/omnibox/releases/tag/v0.1.29

OmniBox AI

1. Switched the Agent used in Pro namespace sharing to Agent 2.1
2. Fixed some known issues

Web

1. Improved the speed of tag generation and title generation
2. Improved task processing stability

WeChat ClawBot

1. Fixed an issue where Q&A in WeChat ClawBot could report an error

Mobile APP

1. Fixed an issue where saving links in the app could cause the web app to crash
2. Fixed incorrect display of the "current namespace recording limit" on the recording page
3. Fixed an issue where parsed file content was visible on the app homepage but missing after opening the card
4. Added a notification center, currently supporting viewing WeChat Assistant replies and OmniBox Community notifications
5. Added automatic clipboard link detection and opening of the "Add Link" dialog
6. Optimized the loading state and button interactions in the "Share to OmniBox" dialog
7. Added "Pricing" and "Renew" entry points
8. Fixed some known issues

v0.1.28 (April 21, 2026)

https://github.com/import-ai/omnibox/releases/tag/v0.1.28

OmniBox AI

1. Basic namespaces now have access to the Agent 2.1 preview trial, with 10 trials per namespace every 24 hours
2. Optimized chat interface interactions and performance
3. Fixed some known issues

Video Parsing

1. Added support for parsing TikTok videos
2. Improved stability of Bilibili video parsing

Web

1. Added a notification center, currently supporting viewing WeChat Assistant replies and OmniBox Community notifications
2. Fixed some styling issues
3. Fixed issue where resources added to a chat dialog were not removed from the dialog reference after being deleted from the resource list
4. Fixed issue where files containing tables were not fully displayed when wide screen mode was enabled

v0.1.27 (April 16, 2026)

https://github.com/import-ai/omnibox/releases/tag/v0.1.27

OmniBox AI

1. Updated Agent 2.1 preview, currently available for Pro namespaces
   • Resource management: browse, search, read, create, edit, rename, move, delete
   • Tag management: browse, search, read, add, remove
   • Frontend operations: expand/collapse folders, locate a specific file

Web

1. Optimized some features and styles on the web
   • Changed "Binding" to "Login Methods" in settings
   • Optimized personal namespace settings and member invitation interface
   • Improved quota display interactions; specific numbers can now be viewed on hover
   • Namespace upgrades are now restricted to namespaces you own
   • Restricted the "Expand" button in Settings > General to owners only
   • Improved the new namespace creation experience
2. Share page optimizations
   • Added background and top action bar when sharing a single file
   • Unified share page style and interactions with the homepage
   • Default expanded folder view when sharing a folder
3. Added support for dragging resources to trash
4. Unified the "Move to Trash" effect between the menu and resource corner
5. Fixed issue where two files stuck together during search
6. Fixed several known issues

v0.1.26 (April 2, 2026)

https://github.com/import-ai/omnibox/releases/tag/v0.1.26

WeChat Assistant

1. Launched the voice-enabled WeChat Assistant (integrated with WeChat ClawBot). Documentation: https://www.omnibox.pro/docs/applications/wechat-clawbot

Video Parsing

1. Optimized Douyin video parsing

Web

1. Fixed the issue where wide screen mode required horizontal scrolling
2. Fixed the style issue of file/folder edit boxes in dark mode
3. Fixed the display issue of long file/folder titles in wide screen mode
4. Optimized the share page file path display to be consistent with the homepage
5. Fixed the issue where some content was missing after sharing
6. Fixed several known issues

v0.1.25 (March 24, 2026)

https://github.com/import-ai/omnibox/releases/tag/v0.1.25

Video Parsing

1. Added support for parsing Douyin videos (image-text content not yet supported)
2. Optimized Xiaohongshu video parsing by removing redundant information

Web

1. Fixed issue where folder icon didn't automatically switch to empty state after clearing folder contents
2. Added character limit for resource titles (files/folders): 128 characters maximum
3. Fixed issue where tags didn't automatically display after file parsing
4. Improved display of Settings > General > Usage interface
5. Improved display of Settings > Background Task Management interface
6. Fixed several known issues

v0.1.24 (March 18, 2026)

https://github.com/import-ai/omnibox/releases/tag/v0.1.24

OmniBox AI

1. Agent 2.0 preview is now available for Pro namespaces
2. Tag generation now follows user's language preference
3. Added error prompts during Q&A
4. Fixed issue with incorrect ordering of citation IDs (Cite ID) in Q&A responses
5. Fixed issue where AI Smart Tag Extraction toggle in Settings > Content was not working

Mobile APP

1. Added support for displaying WeChat Assistant replies via notifications on iOS and some Android devices
2. Added support for saving web pages and other APP content via the share menu
3. Added support for sharing files via the share menu
4. Fixed issue where sharing from Bilibili iOS app to OmniBox was not working

WeChat Assistant, QQ Assistant

1. Added support for collecting international websites

Video Parsing

1. Added support for Xiaohongshu videos

Web

1. Improved display of collection titles for WeChat/QQ Assistant
2. Improved title display when video collection fails
3. Added full title display on hover for left sidebar files
4. Fixed API creation button styling issue in light mode
5. Fixed search result styling in search box (icon, title, content summary)
6. Fixed email icon display issue in Safari browser
7. Fixed issue where expired namespaces still showed expired status after renewal
8. Fixed several known issues

v0.1.23 (March 10, 2026)

https://github.com/import-ai/omnibox/releases/tag/v0.1.23

Web

1. Added namespace version display on homepage and settings page
2. Fixed issue where email content still showed "invite to join namespace" when inviting to collaborate on a specific file
3. Added scrollbar to "Permissions" page
4. Fixed button styling issues on the assistant binding interface
5. Fine-tuned dropdown menu options
6. Optimized visual appearance when entering verification code in dark mode
7. Fixed several known issues

Browser Extension

1. Fixed several known issues

WeChat Assistant, QQ Assistant

1. Fixed issue where Pro namespace users couldn't upload 40 MB files through the assistants

Website

1. Added renewal feature to pricing page: https://www.omnibox.pro/en/pricing/
2. Optimized pricing page
3. User avatar displayed in top-right corner when logged in

Documentation

1. Added pricing content: https://www.omnibox.pro/docs/pricing

v0.1.22 (February 7, 2026)

https://github.com/import-ai/omnibox/releases/tag/v0.1.22

Web

1. Added support for commercial features
2. Added display of personal and team storage usage
3. Added right-click context menu support for "Personal" and "Team"
4. Fixed issue where share expiration time could be set to a past time
5. Fixed issue where share page title didn't follow the file title
6. Improved display when share link has expired
7. Added support for resizing sidebar by dragging left/right on shared pages
8. Added welcome message on shared pages when AI chat is enabled
9. Optimized icons on share pages
10. Fixed issue where settings page appeared empty after binding Google/WeChat and returning to homepage

Browser Extension

1. Improved display in some error scenarios
2. Fixed several known issues

Website

1. Added Pricing page - choose a plan that suits you

Other

WeChat stickers and red packet covers need to be accessed by copying the link and opening it within WeChat.

1. Released OmniBox Series 1 and Series 2 (Spring Festival) stickers on WeChat
2. Added WeChat Red Packet Cover
3. Added OmniBox Community (currently in beta — welcome to join the community to discuss and exchange ideas about product-related topics)

v0.1.21 (January 29, 2026)

https://github.com/import-ai/omnibox/releases/tag/v0.1.21

Web

1. Fixed issue where clipping certain web pages would produce duplicate results
2. Fixed issue where Google Chrome would auto-translate omnibox.pro
3. Fixed several known issues

Browser Extension

1. Improved behavior when extension cannot be used on unsupported pages
2. Improved extension styling in dark mode
3. Fixed several known issues

OmniBox AI

1. Improved wording when referencing local files in conversation responses

QQ Assistant

Pending platform approval

1. Added voice-enabled QQ Assistant (see documentation: QQ Assistant)

Docs

1. Added image parsing documentation: Image Parsing
2. Improved WeChat Assistant documentation: WeChat Assistant
3. Added QQ Assistant documentation: QQ Assistant
4. Improved documentation structure

v0.1.20 (January 22, 2026)

https://github.com/import-ai/omnibox/releases/tag/v0.1.20

Web

1. Fixed issue where shared pages could not be scrolled up and down
2. Fixed scrolling issues on some mobile browsers
3. Fixed several known issues

v0.1.19 (January 16, 2026)

https://github.com/import-ai/omnibox/releases/tag/v0.1.19

Web

1. Added phone number signup and login functionality (currently only supports phone numbers from mainland China)
2. Added Android and iOS PWA (Progressive Web App) support (HarmonyOS does not support PWA)
3. Fixed several known issues

Browser Extension

1. Fixed issue where extension required re-login after logging in under certain scenarios
2. Fixed error when clicking "Open Space" while not logged in
3. Fixed errors when clipping certain web articles with the extension
4. Fixed several known issues

v0.1.18 (January 8, 2026)

https://github.com/import-ai/omnibox/releases/tag/v0.1.18

Web

1. Added Trash feature (resources in Trash older than 7 days will be automatically deleted)
2. Fixed several known issues

Browser Extension

1. OmniBox extension is now available on the App Store, search for OmniBox to install (can be used in Safari browser on macOS)
2. Fixed several known issues

WeChat Assistant

1. Good news: OmniBox Assistant is back! (You need to re-add and re-bind, see documentation: WeChat Assistant)
   • Bad news: It can no longer reply. QAQ
2. Fixed several known issues

Help Docs

1. Added usage instructions for some common features
2. Optimized documentation layout and structure

v0.1.17 (December 31, 2025)

https://github.com/import-ai/omnibox/releases/tag/v0.1.17

Web

1. Fixed issue where unsupported email domains could still register
2. Added supported email domains
   • Currently supported: gmail.com, outlook.com, hotmail.com, 163.com, qq.com, 126.com, foxmail.com, yeah.net, sina.com, yahoo.com, sohu.com
   • If your account uses an unsupported email:
     • If you can still log in, please change your email to avoid being locked out in the future
     • If you cannot log in, please contact us with a supported email and we will manually update it
3. Fixed issue where images were not saved when clipping certain websites
4. Fixed issue where members already in a namespace could be invited again
5. Fixed issue where existing email could be used when changing email
6. Fixed issue where non-existent groups could still be successfully invited to files/folders in team namespaces
7. Fixed several known issues

Browser Extension

1. Optimized the position where the floating window appears when triggered
2. Fixed issue where the extension affected double-click and triple-click text selection
3. Added prompt message when clicking extension on browser built-in pages or omnibox.pro
4. Fixed several known issues

v0.1.16 (December 23, 2025)

https://github.com/import-ai/omnibox/releases/tag/v0.1.16

Web

1. Added Apple ID sign up, login, and account binding functionality
2. Added ability to delete namespace, leave namespace, and delete account
3. Added "Admin" member role
4. Added "Timeout" status in background task management
5. Fixed issue where left resource list didn't update with right resource changes
6. Fixed several known issues

Help Docs

1. Added some frequently asked questions
2. Added offline browser extension installation tutorial

v0.1.15 (December 16, 2025)

https://github.com/import-ai/omnibox/releases/tag/v0.1.15

Web

1. Added ability to set folder name directly when creating new folder
2. Optimized resource tag display
3. Optimized task display after file upload triggers parsing
4. Optimized image loading animation
5. Optimized settings interface style
6. Optimized color settings to follow system, but still requires manual refresh
7. Fixed popup issue during Google binding
8. Fixed several known issues

WeChat Assistant

1. Fixed error when sent file names contain "#xxx"
2. Fixed several known issues

Browser Extension

1. Fixed issue where clicking extension on OmniBox product homepage caused automatic refresh
2. Fixed issue where content couldn't be saved after selection when "selection menu" is disabled but "area selection" is enabled
3. Fixed several known issues

Website

1. For Google Chrome users visiting the website, if they cannot access the Chrome Web Store, clicking "Add to Browser" in the top right will automatically redirect to the browser extension installation documentation page

v0.1.14 (December 8, 2025)

https://github.com/import-ai/omnibox/releases/tag/v0.1.14

Web

1. Fixed parsing failures for some PDF files
2. Fixed issue where locally uploaded .wav files were not being parsed
3. Added image upload functionality (supports .jpg, .jpeg, .png formats)
   • If it's a document: It will go through the document parsing process and be saved as a parsed document
   • If it's not: The image will be summarized and saved as a "document containing only one image"
4. Improved stability when parsing multiple PDF files simultaneously
5. Fixed issue where images from some GitHub sites were not displaying properly when saved
6. Fixed issue where some WeChat official account content failed to save
7. Improved folder/file renaming interaction
8. Improved right-click interaction for folders/files and the "···" button interaction
9. Fixed parsing of Zhihu "Thoughts"
10. Fixed several known issues

WeChat Assistant

1. Fixed issue where WeChat Assistant couldn't save Zhihu "Thoughts"
2. Improved WeChat Assistant's response mechanism for user messages
3. Optimized resource scheduling - when resources are tight or multiple messages are sent in succession, the assistant will process content gradually
4. Added message processing frequency limit - the system only processes one message per user every 3 seconds
5. Added image support for WeChat Assistant
6. Fixed several known issues

Help Docs

1. Updated browser extension installation package for local download
2. Updated "supported file types"
3. Added FAQ entries for "local deployment"
4. Added many new FAQ entries

Browser Extension

1. Improved visual effects for shortcuts in the settings interface
2. Added dark mode

v0.1.13 (November 28, 2025)

https://github.com/import-ai/omnibox/releases/tag/v0.1.13

Web

1. Optimized video parsing
2. Renamed “Browser” to “Extension” on the homepage
3. Added an "Applications" page in "Settings" for users with the Member role
4. Added a character limit for parsed text from uploaded files. If the output exceeds 32,768 characters, an error is returned
5. Fixed parsing failures for some .txt files
6. Fixed parsing failures for some .ppt and .pptx files
7. Fixed parsing failures for some .doc and .docx files
8. Fixed several known issues and improved overall stability

OmniBox AI

1. Improved performance for new-file indexing

Browser Extension

1. Fixed an issue where images were not saved on certain sites when clipping mixed text-and-image content
2. Fixed image layout issues on certain sites when clipping mixed content
3. Fixed general content layout issues on certain sites
4. Improved handling of mixed text-and-image content on certain sites
5. Fixed an issue where vertical videos on some mainland-China sites were not saved
6. Improved handling of horizontal videos on some mainland-China sites
7. Fixed an issue where images in X (Twitter) posts were not saved when clipping

WeChat Assistant

1. Fixed an issue where sending files to WeChat Assistant could fail
2. Fixed an issue where sent videos were not saved
3. Improved reliability when sending certain types of content
4. Added Q&A: the assistant now auto-triggers Q&A or Quick Note based on the message content
   • 4.1 Use ? to force Q&A (e.g., "? Who are you")
   • 4.2 Use ! to force Quick Note (e.g., "! Who are you")

Help Docs

1. Improved documentation for file-upload shortcuts

v0.1.12 (November 20, 2025)

https://github.com/import-ai/omnibox/releases/tag/v0.1.12

Web

1. Added page limit for uploaded PDFs, with a single PDF limited to 32 pages (because some users uploaded 4000-page files, which was really too much to handle)
2. Added duration limit for audio and video files (including web pages), with a single audio or video file limited to 20 minutes (some users uploaded 5-hour videos, which was also really too much to handle)
3. Fixed several known issues

Browser Extension

1. Extension has been adapted for Safari browser (currently under App Store review)
2. Improved first-time user experience (when users install and use the extension for the first time, the page is automatically refreshed to make the extension work on existing web pages)
3. Changed "Currently Visiting" on the extension settings page from auto-save to manual save
4. Fixed several known issues

Other

1. We've added some limits in this release. After analyzing the distribution of online usage, we've tried our best to ensure these limits can meet the needs of the vast majority of users. If this has a significant impact on you, please contact the group admin to discuss

v0.1.11 (November 18, 2025)

https://github.com/import-ai/omnibox/releases/tag/v0.1.11

Web

1. Added automatic document saving to local cache (reliability not guaranteed)
2. Fixed the issue where new files couldn't be created on touchscreen devices like phones and tablets
3. Added confirmation dialog when deleting files and folders
4. Added character limit for individual tags (maximum 20 characters)
5. Fixed the issue where .md files couldn't be uploaded on some systems
6. Optimized the settings page for sharing functionality
7. Fixed several other known issues

OmniBox AI

1. Fixed the issue where "cursor not at the end during conversation, unable to insert line breaks in the middle"
2. Fixed the chat box overflow issue when entering content in a chat with existing messages

Help Docs

1. Optimized the docs for local deployment
2. Updated API docs
3. Added change log: https://www.omnibox.pro/docs/change-log
4. Added QQ group and Jike homepage to contact information

v0.1.10 (November 11, 2025)

https://github.com/import-ai/omnibox/releases/tag/v0.1.10

Web

1. Added file size limit, with a single file size limit of 20MB (because some users uploaded 3GB files, which was really too much to handle)
2. Added email verification code login, simplifying the registration process (SMS login is on the way)
3. Fixed errors when saving model responses as documents
4. Fixed several known issues

v0.1.9 (November 5, 2025)

https://github.com/import-ai/omnibox/releases/tag/v0.1.9

Web

1. Added upload progress display for batch file uploads
2. Support for disabling automatic tagging feature in "Settings / Content / AI Smart Tag Extraction" (name tentative)
3. Fixed possible failures when batch uploading files
4. Fixed the issue where a file might appear twice when dragging it into a folder
5. Fixed parsing errors caused by file encoding
6. Removed the "automatically create Uncategorized when creating new namespace" feature, web bookmarks are saved to "Personal" by default
7. Fixed several other known issues

OmniBox AI

1. Fixed title display issue during conversations
2. Fixed the issue where titles weren't automatically generated during conversations
3. Fixed the issue where English tags might be generated in Chinese input mode

v0.1.8 (November 3, 2025)

https://github.com/import-ai/omnibox/releases/tag/v0.1.8

Web

1. Significantly optimized file upload and download speeds
2. Support for exporting documents as Markdown format
3. Fixed several known issues and improved overall stability

OmniBox AI

1. Support for message retry, editing, and branch switching
2. Optimized long text editing experience in the input box
3. Support for saving model replies as documents with one click

Browser Extension

1. Support for "quick access to current namespace"