Proactive Organization & AI-Based Suggestions

Problem: As content grows, users struggle to keep their workspaces organized. ABLE currently relies on manual structure building, which leads to clutter, duplicates, and lost knowledge. There's no system-level assistant that proactively helps organize or suggests useful actions. Proposed Solution: Introduce a Proactive AI Assistant that suggests relevant organizational or content-related actions, such as: Suggested Actions: “Move this page to Folder: Research?” “Link this file to View: Q3 Planning?” “This highlight matches Topic: AI. Create a summary page?” “You have 3 web tabs about the same domain — group them?” Organizational Intelligence: Detect similar content and suggest grouping Flag duplicate or orphaned items (unused pages/files) Identify outdated items for archiving Workflow Boosters: Recommend new views (e.g. “View: All notes tagged #UX this week”) Auto-build dashboards based on recent patterns Suggest recurring structure from user behavior

File Restoration After Deletion

Add the ability to restore files in the database after they have been deleted.

Flexible Tab Sorting Options

Tabs can only be sorted in one way, limiting organization and workflow flexibility. It’s inconvenient to find and organize tabs when working with many pages, reducing speed and control. Suggest a solution Add multiple tab sorting options: by last opened, by time, by name, by website/domain, and allow custom/manual sorting. Enable users to switch sorting modes as needed.

Show offline notification

Problem Currently, when the user goes offline, there’s no clear indication in the UI. This creates confusion: users may think ABLE is broken or their actions are not working, when in fact the issue is connectivity. Expected Show a toast notification when the app detects offline status (e.g., “You’re offline.”). Toast should auto-dismiss when the connection is back online (or show a brief “Back online” confirmation).

Rethink Vaults

Now ABLE limits the work to only one folder (Vault). This is inconvenient because: You cannot connect multiple folders at the same time (for example, a work project in one folder, personal notes in another, and access to a folder on Google Drive). If the file is opened outside of the current Vault (for example, a .md document in the Downloads folder), double-click on . md file opens ABLE, but the file itself is not loaded. Why is this a problem for you? The user has to either manually move the files into the active Vault, or change the default editor to another one in order to properly open . md files. Loss of natural scenario: users expect that any file opened via ABLE by default will be available for editing — regardless of its location. Suggest a solution Ability to add multiple folders (multi-vault support) and work with them simultaneously or give full access to drive. Implement "Open External File" mode: If a file is opened from outside the current Vault, ABLE should load it in read/write mode, offering to save to an active Vault or continue working "as is".

Image-to-Text (OCR for Extracting Text from Images)

Problem: Users often add screenshots, photos, or scanned documents that contain valuable text. Currently, there is no way to extract or search the text within these images, which limits their usability and discoverability. Proposed Solution: Add built-in OCR (Optical Character Recognition) functionality to extract text from images. Auto-run OCR on image upload Make image text searchable in global or contextual search

Content Auto-Collecting

Problem: Currently, collecting relevant content across a workspace requires manual linking or tagging. This limits discoverability and increases overhead when trying to gather all related notes, highlights, links, or files — especially in dynamic, fast-growing vaults. Proposed Solution: Introduce auto-collecting views or blocks that dynamically pull in related content based on: Search queries / filters (like in Tana): Example: #meeting AND author:Anna Saved as smart blocks or auto-updating sections within a page or view Linked / unlinked reference detection (like in Roam): Auto-detect mentions of a page or keyword, even if not manually linked Display them under: “Linked Mentions” — direct references “Unlinked Mentions” — plain-text mentions, with option to link

Recent Searches

Problem: Currently, there's no way to access or reuse recently entered search queries.

Unify “Link To” and “Move To” Actions for Folders and Files

Problem: In the current implementation, “Link To” and “Move To” appear as separate actions in the context menu for files and folders. This causes confusion for users, who often don’t immediately understand the difference between the two, leading to hesitation and friction: “What’s the difference?” “Which one should I choose?” “What does each one actually do?”

Reuse Existing Tabs in Active Context Instead of Opening Duplicates

Problem: When a user opens an item (e.g. a page, file, or link) that is already present in the active context (e.g. in the sidebar or a view), ABLE still opens it as a new tab instead of reusing the existing one. This leads to: Duplicate tabs of the same item Fragmented workspace Loss of spatial memory / focus Use Cases: From search: I search for a page that is already added to my current context. The app should propose an action activate the existing one. From a pane (library, context screen etc.): I click on something that’s already been added to the context. ABLE opens it again in a new tab — when it should just bring me to the one I already have. Proposed Behavior: If the target item is already open or saved within the current context, then: Focus the existing tab in the sidebar Scroll to and highlight it if it’s out of view Only open a new tab if the item is not present in the current context. Why it matters: Keeps the workspace clean and consistent. Respects user’s intentional organization of their context.

Sidebar views should be displayed as collapsible folders like in Library tree, not as wide panels

Explain the problem as you see it: Currently, views like Landing page, Process, or Read are rendered in the sidebar as wide panels that take up half of the height, instead of being shown as structured collapsible folders, like in the Library. Why is this a problem for you? It breaks the visual and spatial consistency of the interface. When I’m navigating information, I expect a compact, nested structure like in Library, where I can quickly collapse or expand sections. Wide panels interrupt my workflow, especially when I’m managing multiple views or switching between them. Suggest a solution: Render all sidebar views in the same tree structure as used in the Library section — with collapsible folders. Reserve full-panel display only for views that contain actual content, not structural navigation. This will make the experience more predictable and keep navigation tidy.

Al to generate tab/website summary based on the tab/website contents

Chat with tab content

Tab Summarization (Auto-Generated Content Metadata)

Problem: Tabs (especially web pages and files) often contain valuable content, but there’s no way to quickly understand or search what a tab is about without opening it and reading manually. This limits the ability to organize, navigate, and retrieve information across large numbers of tabs.

Link Objects to Multiple Folders

Allow linking objects (folders, pages, PDFs) to multiple folders simultaneously.

Daily notes (base on calendar and md, locally stored)

Implement basic calendar function

Using User-Selected Sources in Context for AI Answers

Currently, AI tools don’t let me collect and provide my own set of sources as context for interactions and research. It limits control and transparency—AI may use irrelevant or untrustworthy materials, and I cannot direct it to rely only on what I consider important. Suggest a solution Allow users to manually select, save, and organize sources (links, PDFs, papers) as context, then instruct the AI to answer questions, write articles, or conduct research specifically based on those curated sources.

Auto-Creation of To-Do List from Notes

Tasks and action items are buried inside meeting notes, emails, and documents. It’s slow and error-prone to manually extract and organize tasks for follow-up. Suggest a solution Automatically detect, extract, and create a to-do list from notes and emails, so actionable items are instantly available.

Redesign pane toolbar

Problem The current pane bar toolbar takes up too much vertical space, which reduces the usable area for content. This feels bulky and distracts from the main workspace. Expected Make the toolbar more compact while keeping all essential actions accessible. Use smaller buttons/icons with clear tooltips. Ensure consistency with the overall minimalistic design of ABLE.

Meeting recording, transcription

Problem Teams and individuals currently need to use external tools to record, transcribe, and extract action items from meetings. This creates fragmentation: meeting knowledge often stays in scattered apps and is not connected to the ongoing work with knowledge base. Expected Recording & Transcription: Capture meeting audio/video, auto-generate a transcript. Speaker Identification: Attribute transcript lines to specific participants. Action Items: Detect and structure decisions, tasks, and action items automatically. Knowledge Integration: Save transcripts and action items directly into ABLE’s knowledge base, linked to relevant contexts/projects.

Keyboard Shortcuts for Context Switching

Frequent context switches require too many clicks, breaking focus and reducing productivity, especially when multitasking or quickly reviewing information. Suggest a solution Add configurable keyboard shortcuts for instant context switching.

If the Library or context screen is active in the pane, pressing CMD+N should create and activate a new object in that pane

Currently, when Library or a context screen is active in a pane, pressing CMD+N does not create and activate a new object directly in that pane. Instead, the behavior can be inconsistent or require additional manual navigation, slowing down the user's workflow. Make CMD+N context-aware: When the Library or a context screen is in active pane, pressing CMD+N should create a new object and immediately activate it in the same pane with active title input. In dual pane as well.

Interface Scaling via Settings Screen

Explain the problem as you see it Currently, interface scaling is not managed through a dedicated settings screen. Why is this a problem for you? Changing interface size is inconvenient; many related settings are spread across different menus. Suggest a solution Add a comprehensive interface scaling option to the main settings screen. Collect existing controls (like AdBlock, updates, camera/microphone permissions, default browser, etc.) and consolidate them—make scaling a fast and visible adjustment alongside other core interface settings.

Basic Table Support in Editor

The editor does not support creating even the simplest tables. It's hard to organize and structure information, compare items, or make data visually clear. Suggest a solution Add basic table creation and editing functionality: ability to add, edit, and remove rows and columns; fill cells with text.

Auto-expand sidebar tree when activating children from Library or Folder Screen

When activating a nested (child) item from the context screen in the pane, the sidebar tree remains collapsed, and the user can’t immediately see the item’s location in the hierarchy. It’s harder to understand context, find the parent chain, and quickly navigate related items. Suggest a solution Automatically expand all parent folders/nodes in the sidebar so the active child item is visible, together with its hierarchy, whenever a nested element is focused from the context screen.

Replace icons across the interface

Problem Current icons feel inconsistent and outdated, which reduces the overall polish of the product. Visual language lacks harmony, making the interface look less professional and harder to scan quickly.

Add “Create View” to Context Menu in Context Screen

There is no “Create View” option when right-clicking on an empty area in the context screen.

New tabs opened in background are hidden when unpinned tabs list full

Problem When the unpinned tabs list is full and scroll position is not in the end of the list, any new tabs opened in the background (Cmd/Ctrl + click) are not visible. This breaks a core navigation workflow and makes it seem like tabs are not opening at all. Expected The tab should open under the tab in which the link was opened and crawl to it. Skol is needed in situations where the last taba on the list and the new taba opened in bekgrunde does not fall into the visible part of the sijbar or in the case of dual-landscape.

Insert/Edit Internal Links to DB objects in Text (Knowledge Base Page)

Enable inserting, adding, and editing internal links to DB objects (files, folders, views) within text on a Knowledge Base page.

Blocks toggling

To allow toggle/hide of the content nested under a given block

Adblock in Browser

Problem: Intrusive ads disrupt browsing, slow down page loading, and clutter the interface. In some cases, they may even pose security risks. Why it’s a problem: Frequent ad interruptions impact my productivity, distract from tasks, and reduce the overall browsing experience. Solution: Introduce an integrated ad-blocker with customizable options, allowing users to block all ads or permit non-intrusive ones. A simple toggle in the settings could enable or disable this feature.

Insert Local Images into Text Blocks (From Disk and Vault)

Problem: Currently, there is no way to insert local images (e.g..png,.jpg) directly into text blocks — whether from the file system or from existing files already stored in the vault. Users are forced to describe visuals with text or manually link to files, which breaks the visual nature of documentation.

Editor rebuild.

After building the PoC with ready-made library, we're going to see how we move to a custom-built editor to allow us more flexibility to implement important future features, like custom blocks, AI etc. We plan to deliver the PoC by the end of current quarter.

Drag & Drop Website from Unpinned Tabs to Inbox Views in Library Screen

Paste Images from Clipboard into Text Blocks

Feature Request: Paste Screenshots (Images from Clipboard) into Text Blocks Problem: Currently, when a user copies an image or takes a screenshot (e.g. with Cmd + Shift + Ctrl + 4 on macOS), pasting it into a text block doesn't insert the image. This breaks expected behavior and forces workarounds like saving the image to disk first, then uploading manually. Proposed Solution: On paste, insert the image block. Upload and store the image in context folder, then embed it in the page.

Preserve Scroll Position and Cursor Focus in Knowledge Base Pages

Problem: When switching between tabs or reopening a page in the knowledge base, ABLE does not retain scroll position or cursor location. This forces users to manually find where they left off, breaking flow and increasing cognitive load — especially on long or structured pages.

Duplicate Tab

Currently, there's no way to duplicate an open tab. If a user wants to view the same item (page, file, site, etc.) in multiple places (e.g. split-view or for reference), they have to manually reopen it, which breaks flow. Proposed Solution: Add a "Duplicate Tab" action, available via: Right-click on tab Hotkey (e.g. Cmd + Shift + D) Command menu

Quick Add Web Link to View

Use Case: As a user, I want to quickly save a web link into my knowledge base (view) without creating a new tab. Proposed Flow: Copy a web link (URL) to clipboard. Open the Link to View popup. Paste the link into the search/input field. Press Enter → the link is instantly added as a new item in the view. Expected Behavior: If the pasted input is a valid URL, it is auto-converted into a web object (with metadata/title fetched if possible). The object is immediately linked to the current view. Optionally, a confirmation or preview is shown after adding. Why it matters: Streamlines knowledge capture from the web. Reduces friction in adding useful references to ongoing work. Matches real-world flow: copy → paste → done.

Auto-expand Folder Contents in Sidebar When Opened as Context

Problem: When a folder is opened as a context, its contents are not automatically expanded in the sidebar. This causes confusion — especially for first-time users — as it's unclear what just happened on the screen. The folder appears to be opened, but no visible change occurs in the UI, leading to disorientation. Proposed Solution: Automatically expand the folder's contents in the sidebar when it’s opened as a context. Why it matters: Makes the structure immediately visible and interactive. Reduces cognitive friction and improves onboarding. Aligns with user expectations for folder-based navigation.

Youtube Full screen

Youtube Fulscreen button doesn’t work. Click - nothing happens.

Sequential File Ordering in Trees

Ensure files in trees are displayed in sequential numeric order (e.g., 01, 02, 03).

Smaller and Auto-Hiding Scrollbars

Current scrollbars in the sidebar and content panes are large and always visible, which visually clutters the workspace. Scrollbars distract from the main content and reduce usable space when working with lists, trees, and documents. Suggest a solution Make scrollbars thinner and automatically hide them when inactive. Show them only on hover or during scrolling, to keep the interface clean and focused.

Shortcuts to call the context menu of selected object

macOS: Control + Enter, Windows: Ctrl+Shift+O — context menu of selected object

Update App Icon (More Contrast & Standout)

The current app icon lacks contrast and doesn’t stand out well among other apps in the dock or taskbar. This makes it harder to quickly recognize ABLE, especially when multitasking. Expected Refresh the app icon design with higher contrast. Ensure it is distinct and recognizable in both light and dark modes. Keep visual consistency with ABLE’s brand identity.

Unified Website Storage with Context Linking

Redesign website storage in the knowledge base to use a single master list of websites, which can then be linked to multiple contexts.

Integrated AI Companion for Daily Notes and Beyond

Explain the problem as you see it: AI tools and chats often have a difficult user interface, requiring constant switching between apps and tedious copy-pasting. This disrupts productivity and limits the potential of AI to act as a true assistant. AI is not just a tool—it should feel like a seamless companion in daily life, helping to organize information, knowledge, and tasks effortlessly. Why is this a problem for you? The current experience feels fragmented. Instead of AI working alongside users as an intuitive helper, it adds friction. Users have to go through extra steps to integrate insights or actions from AI into their workflows, which wastes time and breaks focus. It also limits the AI’s potential as a truly integrated partner in journaling, task management, and knowledge building. Suggest a solution: Reimagine AI as a fully integrated companion that works across all aspects of daily life and productivity. Key functionalities could include: • Contextual Awareness: AI should understand and stay connected with the user’s ongoing tasks, notes, and projects, offering relevant suggestions in real time. • Centralized Integration: AI should be embedded into tools for journaling, daily planning, and knowledge management, removing the need for manual copy-pasting or app-switching. • Seamless Assistance: AI could proactively organize and connect information, set reminders, generate summaries, and even suggest improvements to the user’s notes or plans. • Personalized Partner: The AI should feel like a true partner—helping with brainstorming, offering insights, and even tracking progress toward goals—all within the same platform. • Enhanced UI/UX: The interface should be user-friendly, intuitive, and designed to make interactions with the AI as natural as possible.

Add triage to tab pin process (simplify and streamline the pin experience)

Dynamic context tree generation

During the session, the system will iteratively analyze the possible context and try to understand how to dynamically build and represent the information structure based on that context (available in the user's KB and maybe the web) to the user.