Frequently Asked Questions

This page answers some of the questions Linkly AI users ask most often.

How does Linkly AI protect user data privacy?

Linkly AI is built on a local-first architecture. Your documents, full-text index, vector index, and the embedding model itself all run on your device — nothing is uploaded to any server by default.

What stays on your device

Document originals: stay in their original folders. Linkly AI only reads them — it does not copy or relocate them.
Full-text index (BM25): built locally with Tantivy.
Vector index: stored in a local database.
Embedding model: a local embedding model runs entirely on your device. Apple Silicon Macs automatically use Metal GPU acceleration, and Windows / Linux use GPU acceleration when there’s a discrete graphics card.
App logs: written to local files only — never auto-uploaded.

Where chat data goes

When the chatbot calls a large language model, where the request goes depends on the provider you choose:

Local model

Ollama, LM Studio, or any other OpenAI-compatible local service. Data stays entirely on your machine.

Linkly Official

Forwarded to a third-party model provider via api.linkly.ai. Requests pass through Linkly’s servers.

Third-party direct

Connects directly to OpenAI, Anthropic, etc. Requests do not pass through Linkly’s servers.

You can add, switch, or disable any provider in Settings → AI Models.

User Experience Improvement Program (telemetry)

To help us understand which features are used and how the app is running, Linkly AI sends an anonymous usage report by default:

What is reported	What is NOT reported
Feature usage counters (aggregated by action)	Document content, file names, paths
App version, OS, architecture	Chat content, search queries
A locally generated random device ID	API keys, custom URLs

You can turn it off any time in Settings → Data Privacy → User Experience Improvement Program. Once disabled, any in-memory events that have not yet been sent are discarded as well.

The “Data Privacy” panel in Settings shows a visual breakdown of where each feature’s data flows (Local / Official cloud / Third-party), so you can see at a glance what does and does not leave your device.

Privacy commitments

No third-party analytics SDKs (Google Analytics, etc.).
We do not read your browser history or clipboard.
No mandatory account login — core features work offline.
App logs stay on your machine. They are sent to us only if you explicitly share them.

How long does Linkly AI take to finish indexing?

Indexing time depends on the number of files, file types, machine performance, and performance mode. Linkly AI indexes in three stages:

Filename quick index (seconds)

As soon as files are discovered, their paths and names are written to the full-text index, so you can search by filename even while content indexing is still running.

Full-text extraction and BM25 index (minutes to hours)

Document contents (txt, md, html, docx, doc, pptx, xlsx, csv, epub, rtf, pdf, images, plus audio and video once transcription is enabled) are parsed; outlines and metadata are written to Tantivy. Multiple workers can run in parallel.

Vector embedding (minutes to hours)

The local embedding model generates a vector for each document chunk and writes it to the vector index.

What affects speed

File count: roughly linear with total time.
File format: plain text is fastest; PDFs require page parsing.
Machine performance: Apple Silicon Macs use Metal GPU acceleration for embeddings, which is significantly faster than CPU inference. On Windows / Linux, embeddings use GPU acceleration when there’s a discrete graphics card, and the CPU otherwise.
Performance mode: pick Performance / Balanced / Auto in the Policy area of the Index management page. Performance mode uses higher concurrency and more CPU; Auto switches to Performance when the computer is idle.

Rough expectations

These are order-of-magnitude estimates only — actual times vary considerably with hardware and file mix:

Scenario	Approximate time
Thousands of plain-text files (txt / md), M-series Mac	A few minutes
Tens of thousands of mixed formats (some PDFs), M-series Mac	Tens of minutes to a few hours
Many images requiring OCR	Significantly longer
Same workload on Windows / Linux without a discrete GPU	Slower than Mac

You can watch progress live (Processed / Pending / Can’t extract) on the Index management page in the main window. While indexing is running, a spinning icon appears in the launcher footer; clicking it also opens that page. Indexing runs in the background and does not block search — once filename indexing is done, you can search.

If you are just trying Linkly AI out, start with a small knowledge base (hundreds to thousands of files), then add larger directories once you are familiar with it.

Does the embedding model or OCR use integrated or discrete graphics?

On Windows and Linux, only a discrete GPU is used; integrated graphics are not. In our tests, integrated graphics ran these models more slowly than the same computer’s CPU, so without a discrete GPU the CPU is used directly. Apple silicon Macs use the GPU built into the chip. Each task uses them differently:

Task	Windows / Linux	Apple silicon Mac
Semantic indexing (embedding model)	Discrete GPU if present, otherwise CPU	GPU
Scanned-document OCR: text detection and recognition	CPU	CPU
Scanned-document OCR: layout analysis	On Windows, discrete GPU if present, otherwise CPU; on Linux, CPU	GPU
Audio and video transcription	CPU	Chip acceleration first, CPU if that fails

A few more points:

OCR layout analysis uses the GPU only for PDFs of 4 pages or more. Single images and short PDFs use the CPU.
With multiple discrete GPUs, semantic indexing uses only the one the system lists first; it doesn’t pick by VRAM size. OCR layout analysis uses the GPU assigned to Linkly AI in Windows graphics settings.
When VRAM runs short, the driver isn’t supported, or the GPU keeps erroring, the app falls back to the CPU automatically and indexing isn’t interrupted.
When semantic indexing switches to the cloud model, the local GPU isn’t involved. Windows and Linux devices on ARM currently use the CPU only.

To stop using the GPU entirely: click the index status icon at the bottom of the main window’s left sidebar to open the Index management page, then turn off Hardware acceleration under Hardware on the Advanced tab.

Windows and Linux: with it off, everything runs on the CPU — slower but steadier.
Mac: this switch only controls whether OCR layout analysis uses the GPU; semantic indexing and transcription are unaffected.

Turning the switch back on makes the app try the GPU again.

Why can’t I find a particular file?

Start with one triage step — search for it by filename. The three possible outcomes point to three completely different problems:

What searching the filename does	What it means	Where to go next
Found, with a grey Not indexed · … badge	The file made it into the index, but its content was skipped	What the badges mean, below
Found, no badge, but content still isn’t searchable	The content hasn’t been extracted yet, or the relevant switch is off	Next section
Not found even by filename	The file never entered the index at all	The section after that

This step matters: the three cases are handled completely differently, and skipping triage is a good way to spend a long time digging in the wrong direction.

Files whose content didn’t make it into the index can all be viewed by reason on the Can’t extract tab of the Index management page; most groups offer Retry all or retrying files one at a time.

Files and folders left out by exclusion rules don’t appear in the library’s file list. The folder they sit in gets a funnel icon after its name: hover over it to see what’s left out, and click it to open the matching rules.

What the badges mean

Badge	What actually happened
Not indexed · cloud placeholder	The file exists in your cloud drive, but hasn’t actually been downloaded to this machine — all that’s on disk is a placeholder
Not indexed · signature mismatch	The file’s real content doesn’t match its extension (say, a file named `.pdf` whose content isn’t a PDF at all)
Not indexed · no audio track	This video has no audio track, so there’s nothing to transcribe
Not indexed · transcription failed	Audio or video transcription couldn’t complete (undecodable, or an unsupported audio format)
Not indexed · language unsupported	The language spoken in the media isn’t covered by the current transcription model; the file itself is fine
Not indexed · extraction failed	Text extraction failed: over the size limit, password-protected, unreadable, or in a text encoding that couldn’t be identified
Not indexed	Skipped for some other reason

Hovering the badge gives you the same explanation. These files stay in your search results; they just have no content to search. Files whose extraction failed are retried automatically once their content changes, and you can also retry them by hand on the Can’t extract tab.

What cloud placeholders are about

iCloud Drive, OneDrive, Dropbox, Synology Drive and similar services all have a files on-demand feature (also called “online-only” or “smart sync”): the file is visible in Finder or File Explorer, but its content still lives in the cloud and is only downloaded when you double-click to open it. Linkly AI deliberately does not open these files — opening one triggers a download, which would mean a single indexing pass could pull hundreds of gigabytes of your cloud drive onto your machine, filling the disk and saturating your bandwidth. So it judges by attributes only: recognise a placeholder, skip the body, keep the filename indexed.

What to do: in your cloud client, mark the directories you actually need to search as “always keep on this device” (each vendor names it differently — OneDrive calls it “Always keep on this device”, Dropbox “Make available offline”, iCloud is a right-click “Download Now”). Once the files land locally, Linkly AI notices they’ve become available on the next scan and indexes the content automatically — you don’t need to rebuild the index or re-add the folder.

Scans are triggered automatically on app start, on configuration changes (adding or removing folders, for example), and when a burst of file changes is detected. If you want one right now, restarting the app is the easiest way.

What to do about a signature mismatch

This means the file header’s signature bytes don’t match the extension. It’s common with stub or shortcut files produced by certain sync tools, half-downloaded files from an interrupted transfer, or files whose extension was renamed by hand. Open it with the corresponding application to check — if it doesn’t open at all, Linkly AI was right to skip it; if the extension is simply wrong, correcting the extension is enough.

Linkly AI’s trade-off here is better to miss than to falsely reject: if reading the file header hits a transient IO error or permission hiccup, it is treated as “content available”. A single momentary failure will never permanently mark a real file as a stub.

Found by filename, no badge, but the content still isn’t searchable

No badge means the file itself is fine — it’s the content extraction step that hasn’t happened, or can’t. In order of likelihood:

The relevant extraction switch is off. Images and scanned PDFs depend on Text recognition (OCR); audio and video depend on Audio transcription and Video transcription — and those last two are off by default. Check in the Management area of the Index management page.
Its turn hasn’t come up yet. OCR and media transcription are the heaviest tasks, so a slow queue is normal. The Management area of the Index management page shows which file they’re currently working on.
A legacy .doc exported on macOS cannot be parsed yet. Files produced by textutil or by the “Word 97” export in Pages use an internal layout that differs from Microsoft Word’s, so they are skipped for now. Re-save them as .docx in Word, or convert them to another supported format.

Index Settings

What every switch does, its default, and what each status means — all on this page.

Not found even by filename

This means the file never entered the index at all. There are five common causes: 1. The folder isn’t being watched. Check in the main window’s Library that its parent directory has been added. If it hasn’t, click Add folder in the Library’s left column to add it. 2. The format isn’t on the supported list. Linkly AI uses an allowlist: only PDF, DOCX, DOC, PPTX, XLSX, CSV, EPUB, RTF, Markdown, TXT, HTML, images, audio, and video extensions are picked up. htm and avi are currently not among them — a file outside the allowlist doesn’t even get its filename registered. The full list is in the Tools Introduction. 3. It’s inside an ignored directory. The following directories, and everything below them, are skipped wholesale:

Hidden directories starting with a dot (.git, .cache, .dropbox.cache and so on)
node_modules and __pycache__ and similar build / dependency caches
NAS sync metadata directories: #recycle, #SynoRecycle, @eaDir (Synology), @Recycle (QNAP)
System directories: $RECYCLE.BIN, System Volume Information
Linkly AI’s own data files: the app data folder ai.linkly.desktop, wherever it is stored

Inside a Git repository, files excluded by .gitignore are likewise not indexed. If your notes happen to sit in a repo’s ignored directory, move them out, or click Add folder in the left column of Library to add that directory on its own. The full set of built-in rules is shown at the bottom of the Rules tab on the Index management page.

4. This type of file is switched off. Types switched off under Index management → Rules → File types (images or video, for example) are not scanned, and files already indexed are removed. 5. An exclusion rule leaves it out. Under Index management → Rules → Exclusion rules, check the rules for All folders and for this folder itself; while you edit, the files each rule leaves out are listed below. Folders with a funnel icon at the end of their row in the library’s left column are the ones with exclusion rules. A .ignore file in the folder can leave files out too. See Index Scope.

Can I add files from a NAS?

Yes. First mount the NAS shared folder in your operating system, then add it to the library just like a local folder:

Windows: use “Map network drive” in File Explorer (e.g. Z:\), or use a network path like \\NAS-name\share-name directly
macOS: in Finder choose “Go → Connect to Server”, enter smb://NAS-address/share-name; once mounted it lives under /Volumes
Linux: mount the SMB / NFS share with your system’s network locations feature or mount

Then click Add folder in the left column of Library in the main window and pick the mounted folder. Folders like this carry an External tag. Things to keep in mind with a NAS:

Indexed content isn’t lost when the NAS disconnects or isn’t mounted. The folder shows Offline in the Library; the app checks every 15 minutes and rescans automatically once it’s back.
This computer may not notice right away when another computer changes files on the NAS. The operating system doesn’t always report changes on network shares promptly, especially on macOS and Linux; missed changes are picked up the next time Linkly AI launches.
The first index is slower than on a local disk, more noticeably the more files there are, depending on the network and the NAS’s read speed.
Synology and QNAP recycle-bin and thumbnail directories (#recycle, @eaDir, and so on) are skipped automatically and never indexed.

Why can’t I search what was said inside my audio and video files?

Because audio and video transcription is off by default. While it’s off, media files are still registered in the index (so filename search finds them), but what was said inside them is never transcribed. Turn on Audio transcription and Video transcription in the Management area of the Index management page. The app downloads the speech recognition model and then works through all your existing media files automatically — no rescan and no index rebuild required.

Index Settings

This page explains what every switch on the Index management page does, its default, and the full troubleshooting path when indexing goes wrong.

How do I keep a conversation focused on a single document?

There are two ways to point at a document:

Start from the document: select the file in Library and click Chat with this file in the toolbar above it, or right-click the file and choose that item. Right-clicking a search result in Find offers it too. This starts a new conversation with the file placed in the composer. Folders and libraries can start one the same way (Chat with this folder, Chat with this library).
Use @ in the composer: type @ in the Ask composer, then a keyword from the filename, and pick the file from the candidates.

Note that pointing at a file applies only to the current message. The composer clears once the message is sent; if a follow-up doesn’t @ the file again, the AI searches your whole library and may cite other documents. Also, pointing at a file is only a hint to the AI, not a hard limit on retrieval — it may still search other material. To keep a whole conversation on that one document:

@ the file again with every follow-up
Say in your question: “Answer only from this document; if it isn’t in the document, just say so”

A file has to finish indexing before it can be @-mentioned. Conversations don’t support uploading attachments, so put any file you want to discuss into your library first.

How many files can a cloud library hold? Is it based on file size?

It’s based on file count, not file size, and there’s no total-capacity limit:

	Free	Pro
Files per library	500	2000
Cloud libraries you can create	1	30

What gets pushed to the cloud isn’t the original file but the text Linkly AI extracts from it (a transcript for audio and video), along with the outline and the index data used for retrieval. So a PDF of several hundred MB and a text file of a few KB each count as one file. Once you hit the limit:

If a push would take the library’s file count over the limit, the whole push is skipped, and the app shows Library file limit reached
Pushes that only modify or delete existing files are unaffected
Files already uploaded are all kept, not locked

Also, when the text extracted from a single document exceeds 10 MB, the cloud doesn’t build a search index for it. The full quota details are in Cloud Libraries.

Can I use the AI features on a machine without internet access (offline / intranet)?

Yes. OCR, speech transcription, and semantic search all run locally (cloud features like the AI chat and Skills installation are the exception) — the only catch is that the models they use are downloaded over the network by default:

The semantic search and OCR models, about 890 MB, are downloaded on first run
The speech transcription model, about 161 MB, is downloaded only once you turn on audio/video transcription

On a machine with no internet access at all, download the installer and the official offline model pack on a connected machine first, carry them over, and install step by step.

Offline Installation Guide

What to download first, which files to copy over, and how to install — the full flow on one page.

What should I do if the Linux AppImage will not start?

Linux distributions vary widely in their base runtime libraries, desktop components, and drivers. An AppImage carries most application dependencies, but it does not replace the operating system’s glibc, C++ runtime, FUSE, or graphics drivers, so it cannot guarantee that it will run on every Linux distribution. Linkly AI ships Linux AppImages for x86_64 and arm64 (aarch64); both require glibc 2.35 or newer. Start by running these commands from the directory containing the AppImage:

uname -m
ldd --version | head -n 1
chmod +x LinklyAI-*.AppImage
./LinklyAI-*.AppImage 2>&1 | tee linkly-ai-linux-startup.log

uname -m prints x86_64 or aarch64; download the AppImage that matches. The last command launches the app while saving terminal errors so you can match them to the cases below.

Permission denied

The AppImage is not executable. Run:

chmod +x LinklyAI-*.AppImage
./LinklyAI-*.AppImage

Do not launch Linkly AI with sudo; doing so can leave configuration and data files that your regular user cannot access later.

FUSE, libfuse.so.2, or AppImage mount errors

AppImages normally mount themselves through FUSE. Install the FUSE 2 compatibility library from your distribution’s official repository only:

Distribution	Command
Ubuntu 22.04 / Debian 12 and derivatives	`sudo apt install libfuse2`
Ubuntu 24.04 or newer	`sudo apt install libfuse2t64`
Fedora	`sudo dnf install fuse-libs`
Arch Linux	`sudo pacman -S fuse2`
openSUSE	`sudo zypper install fuse libfuse2`

If FUSE is unavailable in your environment, you can temporarily extract and run the AppImage:

./LinklyAI-*.AppImage --appimage-extract-and-run

This is slower and bypasses FUSE only; it does not fix GLIBC, CPU architecture, or driver incompatibilities. See the official AppImage FUSE troubleshooting guide for background.

GLIBC_x.y, GLIBCXX_x.y, or CXXABI_x.y not found

The operating system’s base runtime is too old, and the failure happens before Linkly AI actually starts:

GLIBC_x.y comes from glibc; the current minimum is glibc 2.35.
GLIBCXX_x.y and CXXABI_x.y come from the system C++ runtime, libstdc++.

First install all official updates offered by your distribution and reboot. If the error remains, upgrade to a newer release supported by that distribution. Installing FUSE or downloading the AppImage again will not change this result.

Do not manually replace /lib/.../libc.so.6, copy glibc or libstdc++ from another distribution, mix repositories from different distributions, or force a downloaded system runtime through LD_LIBRARY_PATH. These workarounds can break the desktop, package manager, or the entire operating system.

Exec format error, or the AppImage does not match uname -m

The package does not match your CPU architecture. Linux AppImages are published for both x86_64 (Linkly AI_<version>_amd64.AppImage) and arm64 (Linkly AI_<version>_aarch64.AppImage); download the one whose name matches uname -m (x86_64 → amd64, aarch64 → aarch64). Changing permissions or installing runtime libraries cannot fix an architecture mismatch.

The process starts, but no window appears or it exits immediately

This is commonly related to the combination of Wayland / X11, Mesa or proprietary graphics drivers, desktop portals, and WebKitGTK. Install the official system and graphics-driver updates offered by your distribution, then launch again from a terminal and keep the output:

./LinklyAI-*.AppImage 2>&1 | tee linkly-ai-linux-startup.log

Avoid copying unknown WebKit, GTK, or graphics environment variables from the internet into permanent system configuration: the same switch can behave differently across WebKitGTK and driver versions. If the app still will not start, send us the following diagnostics together with linkly-ai-linux-startup.log:

cat /etc/os-release
uname -a
echo "session=$XDG_SESSION_TYPE desktop=$XDG_CURRENT_DESKTOP"

The app runs, but its built-in update fails

Updating an AppImage requires replacing the current AppImage file. Keep it in a directory writable by your user, such as ~/Applications, rather than a read-only mount or a root-owned system directory:

mkdir -p ~/Applications
mv LinklyAI-*.AppImage ~/Applications/
chmod +x ~/Applications/LinklyAI-*.AppImage

Documentation can solve missing execute permission or FUSE, but it cannot change the GLIBC or CPU architecture required by the binary. If the app never creates app.log, use the terminal output saved in linkly-ai-linux-startup.log instead.

Is there a Linux server or Docker version?

Not yet — we’re preparing one. If you have ideas or needs around this, or would like early access, get in touch.

Contact us

Leave a note in GitHub Issues, or reach us through any channel on the community page.

Until then, you can install the Linkly AI CLI on the server or in the container and access the Linkly AI desktop app on another computer over the LAN or remotely, as described in the next section.

How do I use the CLI to access Linkly AI on another computer?

The Linkly AI CLI connects only to the Linkly AI on the same machine by default. There are two ways to reach another computer, LAN and remote, and both require CLI 1.0.0 or later (check with linkly --version, upgrade with linkly self-update).

LAN (free)
Remote (Pro)

Use this when both computers are on the same network.

Turn on LAN access on the computer being accessed

Click the Agent icon at the bottom of the main window’s left sidebar to open the Agent usage page. Under Management, turn on Allow LAN access; Allow local access must stay on too. A token is generated automatically the first time you turn it on.

Copy the address and token

On the same page, click View MCP connection settings in the Integration section and copy the address (like http://192.168.1.10:60606/mcp) and token from the LAN section.

Save the connection on the computer running the CLI

Run this yourself, in your own terminal — don’t have an AI assistant run it:

linkly auth set-key <token> --lan --endpoint http://192.168.1.10:60606

You only need to save it once; switching terminals or agent sessions doesn’t require entering the token again.

Add --lan when you use it

linkly search "quarterly report" --lan

Use linkly doctor --lan to check that the connection works.

Use this when the two computers aren’t on the same network; they connect through Linkly AI’s secure tunnel.

Turn on remote access on the computer being accessed

Sign in to your Linkly AI account and turn on Allow remote access (internet) under Management on the Agent usage page. This feature requires a Pro plan, and Linkly AI on this computer has to keep running.

linkly auth login

Follow the prompt to confirm in your browser.

Add --remote when you use it

linkly search "quarterly report" --remote

Keep in mind:

The CLI can save only one LAN connection; running set-key --lan again overwrites the previous one. To reach another computer temporarily, add --endpoint <address> --token <token> to the command — this isn’t saved.
Saved connections are stored in ~/.linkly/credentials.json.
LAN access is best turned on only in trusted networks such as home or office. Don’t paste the token into any AI conversation — conversations are sent to the model provider and kept in the chat history. If a token leaks or appears in a conversation, regenerate it under View MCP connection settings; the old token stops working immediately.
Agents that use the CLI through Skills (such as Claude Code) pick up the saved connection automatically; this needs CLI 1.0.0 and Skills 0.9.0 or later.

Where do I find the settings I need?

To open Settings, click the account icon in the bottom-left corner of the main window and choose Settings from the menu, or press ⌘ + , (Ctrl + , on Windows/Linux). Settings is split into pages by function; these are the ones used most:

Language: the interface display language
Launch at Login: start Linkly AI when you log in to your system
Silent Launch: when started at login, stay in the background instead of opening the main window — on by default, so indexing and the MCP server run without getting in your way
Linkly AI Space: the folder that’s indexed by default — you can open it or move it elsewhere; see Linkly AI Space
App Data: where the index, local models, settings, and logs are stored — click Open to go straight there (that’s where you grab the log); you can also move it to another disk, and the app will restart and migrate the data; see How do I move app data to another disk

Settings → Appearance: theme, font size, Dock icon

Theme: Light / Dark / System
Font Size: Small / Standard / Large, applied to both the interface and chat content
Show Dock Icon (macOS): turn it off and the app lives only in the menu bar tray, taking up no space in the Dock

Settings → About: version, updates

Check for Updates: manually check for and download a new version; when it’s downloaded, click Restart to Update
Join Beta Program: get beta updates and early access to new features, at the cost of possibly running into less thoroughly tested changes
From here you can also jump to the changelog, privacy policy, terms of service, and the CLI and Skills documentation

Settings → Shortcuts: global shortcuts and the full shortcut list

Global shortcuts: Search toggles the Launcher, and Main window opens the main window. Click once to re-record a shortcut, or use Reset to default or Clear. If it says the combination is already in use or reserved by the system, pick another one
Shortcut list: lists every shortcut in the app. Pressing ⌘ + / (Ctrl + / on Windows/Linux) in the Launcher opens this page directly

Pause Indexing is not in Settings. Click Pause on the first card of the status area on the Index management page in the main window, or right-click the L icon in the tray and choose Pause Indexing. See Index Settings for details.

Can I change where the Linkly AI Space is?

Yes. The Linkly AI Space is the default folder for notes, clippings, and other material. To change it:

Go to Settings → General and find the Linkly AI Space row
Click the Change icon button on the right, then click Continue in the confirmation dialog
Pick a folder; Linkly AI creates a LinklyAI folder inside it as the new space. If the folder you pick is itself named LinklyAI, it’s used as-is

Changing it doesn’t move existing files: files at the old location stay where they are and keep being indexed; new notes and clippings go to the new location from then on. To bring old files over, move them yourself. To go back to the original location, click Restore Default next to it.

If you see The current folder is a system folder. Pick a different folder to migrate to. while changing it, you’re changing App Data in the row below, not the Linkly AI Space. App Data can’t go in a system directory such as Windows; pick an ordinary folder instead — see the next section.

How do I move app data to another disk?

App Data is the folder where Linkly AI keeps its database, index, local models, settings, and logs, and with lots of material it can take up several GB. By default it’s on the system drive, and it isn’t the same thing as the software’s install directory: even if you installed to drive D, app data is still in the default location below.

System	Default location
Windows	`%APPDATA%\ai.linkly.desktop`
macOS	`~/Library/Application Support/ai.linkly.desktop`
Linux	`~/.local/share/ai.linkly.desktop`

When the system drive is running low on space, you can move the whole app data folder to another disk from Settings. The Linkly AI Space is a separate setting and doesn’t move along with app data.

Steps

Open the App Data setting

Go to Settings → General and find the App Data row.

Choose a new location

Click the Change button on the right (the folder icon with a pen), click Change and restart in the confirmation dialog, then pick a folder. Linkly AI creates LinklyAI\ai.linkly.desktop inside it for the data. For example, pick D:\Linkly AI Data and the data goes to D:\Linkly AI Data\LinklyAI\ai.linkly.desktop.

Wait for the move to finish

After the app restarts it shows a progress window: it copies the data, verifies the copy, switches to the new location, and restarts automatically when done. With lots of material this can take several minutes — keep the window open and don’t shut down meanwhile.

Check the result

Back in Settings → General → App Data, the new path should be shown, and the size next to it is how much space the data actually takes.

The new location has to meet these conditions; if it doesn’t, you’re told why:

It’s a folder, not the root of a drive (such as D:\)
It isn’t a system directory (such as the Windows folder), a network drive, or a shared folder; folders under Program Files are fine
It doesn’t overlap the Linkly AI Space or any watched folder, and isn’t inside the current app data folder
It has free space of at least 1.1 times the current app data size
It doesn’t already contain a non-empty LinklyAI\ai.linkly.desktop
It can’t be moved while a model is downloading; wait for the download to finish and try again

Why is there still a folder at the old location after the move?

Once the move finishes, the data at the old location is deleted, and a link with the same name is left in its place, pointing to the new location (a directory junction on Windows, a symbolic link on macOS and Linux). Linkly AI finds the data at the new location through this link every time it launches, so the link has to stay.

The link takes up no space. If you check its Properties in File Explorer, it shows the same size as the new location, because the system counts the data the link points to. To confirm the space was freed, look at the free space of the system drive under “This PC”: after the move it should have grown by the size of the app data.
The link and the new location look exactly the same, with content, size, and modified times changing together. That’s because opening the link shows you the data at the new location — it isn’t two copies syncing with each other.
On Windows, %LOCALAPPDATA%\ai.linkly.desktop holds the app interface’s cache. It isn’t part of the move, and it’s rebuilt automatically if deleted.

Don’t delete the link at the old location, and don’t open it to delete the files inside. Deleting the files inside deletes the real data at the new location. Deleting the link itself makes Linkly AI treat its next launch as a fresh install: it downloads the models again and builds the index from scratch, and the data at the new location is no longer used.

I deleted the link by accident

If you haven’t launched Linkly AI since deleting the link, recreate it with the command below (replace the path with your new location):

Windows (Command Prompt):

mklink /J "%APPDATA%\ai.linkly.desktop" "D:\Linkly AI Data\LinklyAI\ai.linkly.desktop"

macOS (Terminal):

ln -s "/Volumes/new-location/LinklyAI/ai.linkly.desktop" ~/Library/Application\ Support/ai.linkly.desktop

Linux (Terminal):

ln -s "/new-location/LinklyAI/ai.linkly.desktop" ~/.local/share/ai.linkly.desktop

If you’ve already launched it since deleting the link, a new, empty data folder will have appeared at the old location. Stop using the app and contact us for help recovering.

I want to move back to the default location, or to a different disk

Back to the default location: click Restore Default (the circular-arrow icon) in the App Data row. Linkly AI restarts and copies the data back to the default location; the system drive needs enough free space. The custom location is cleared only after the copy finishes.
To a different disk: click Change again and pick the new location following the steps above.

Will I lose data if the move is interrupted or fails?

No. Until the copy at the new location has been verified and the app has started normally from there, the original data stays exactly as it was. If the move is interrupted (by a power cut or a force quit, for example), a recovery window appears on the next launch; click Retry to continue, or Cancel the change and keep the previous location. If you run into something you can’t resolve, click Open log and send us the log.

Can I move it to an external drive?

Yes, but the drive has to be connected every time before you use Linkly AI. If it isn’t, launching shows Your app data folder can’t be reached; connect the drive and click Retry. We don’t recommend putting app data on a portable drive you plug and unplug often.

What happens when I uninstall Linkly AI?

On Windows, if you tick Delete the application data when uninstalling, the data moved to the new location is deleted too. If you only want to uninstall and reinstall while keeping your data, leave it unticked; after reinstalling, Linkly AI finds the original data automatically through the link.

How do I get Linkly AI’s application runtime logs?

While running, the app automatically writes its full log to a local app.log file (capped at 10 MB per file, with rotation). Attaching this file when you report an issue dramatically speeds up debugging. Sensitive data is already redacted. There are two ways to grab the log:

Option 1: Open from inside the app (recommended)

This is the simplest path:

Open the General page

In Linkly AI, go to Settings → General.

Open the app data directory

Find the App Data row and click the Open button on the right. Your file manager will open at the folder that holds the app’s data.

Grab app.log from the logs subfolder

Inside that folder, open the logs/ subdirectory and send us app.log.

Option 2: Open the app data directory manually

If the app has crashed or won’t launch, open the folder directly from disk:

macOS
Windows
Linux

Open Finder.
From the menu bar, choose Go → Go to Folder… (or press ⌘ + ⇧ + G).

Paste this path and press Return:

~/Library/Application Support/ai.linkly.desktop/logs

Find app.log in that folder and send it to us.

Press Win + R to open the “Run” dialog.
Paste this path and press Enter:
%APPDATA%\ai.linkly.desktop\logs
Find app.log in that folder and send it to us.

In a terminal:

xdg-open ~/.local/share/ai.linkly.desktop/logs

Or copy app.log directly:

cp ~/.local/share/ai.linkly.desktop/logs/app.log ~/Desktop/

What if the app crashes during startup?

If the app crashes immediately on launch, app.log may not have been written yet. In that case the crash information is saved separately to a startup-panic-<number>.json file in the linkly-ai-diagnostics diagnostics folder. The next time you open Linkly AI, a Linkly AI didn’t shut down properly prompt appears, and this crash information goes into the diagnostic report along with it. Write a line about what you were doing before the crash and click Send report. If the app crashes on every launch and you never see that prompt, you can find the file manually at the path below:

macOS
Windows
Linux

Open Finder.
From the top menu choose Go → Go to Folder… (or press ⌘ + ⇧ + G).

Paste the path below and press Enter:

~/Library/Application Support/linkly-ai-diagnostics

Find the .json file starting with startup-panic- in that folder and send it to us.

Press Win + R to open the “Run” dialog.
Paste the path below and press Enter:
%LOCALAPPDATA%\linkly-ai-diagnostics
Find the .json file starting with startup-panic- in that folder and send it to us.

Note that this path uses %LOCALAPPDATA% (local data), not the %APPDATA% (roaming data) used by app.log.

In a terminal:

cp ~/.local/share/linkly-ai-diagnostics/startup-panic-*.json ~/Desktop/

This file is only created when a crash happens during startup, and it’s cleaned up automatically after the next normal launch puts it into a diagnostic report. If this file does not exist either, take a screenshot of the crash dialog or window and send it along with your OS version.

Where to send it

Open a GitHub issue, or reach us through any channel on the community page.

​How does Linkly AI protect user data privacy?

​What stays on your device

​Where chat data goes

Local model

Linkly Official

Third-party direct

​User Experience Improvement Program (telemetry)

​Privacy commitments

​How long does Linkly AI take to finish indexing?

​What affects speed

​Rough expectations

​Does the embedding model or OCR use integrated or discrete graphics?

​Why can’t I find a particular file?

​What the badges mean

​What cloud placeholders are about

​What to do about a signature mismatch

​Found by filename, no badge, but the content still isn’t searchable

Index Settings

​Not found even by filename

​Can I add files from a NAS?

​Why can’t I search what was said inside my audio and video files?

Index Settings

​How do I keep a conversation focused on a single document?

​How many files can a cloud library hold? Is it based on file size?

​Can I use the AI features on a machine without internet access (offline / intranet)?

Offline Installation Guide

​What should I do if the Linux AppImage will not start?

​Is there a Linux server or Docker version?

Contact us

​How do I use the CLI to access Linkly AI on another computer?

​Where do I find the settings I need?

​Can I change where the Linkly AI Space is?

​How do I move app data to another disk?

​Steps

​Why is there still a folder at the old location after the move?

​How do I get Linkly AI’s application runtime logs?

​Option 1: Open from inside the app (recommended)

​Option 2: Open the app data directory manually

​What if the app crashes during startup?

Where to send it

How does Linkly AI protect user data privacy?

What stays on your device

Where chat data goes

User Experience Improvement Program (telemetry)

Privacy commitments

How long does Linkly AI take to finish indexing?

What affects speed

Rough expectations

Does the embedding model or OCR use integrated or discrete graphics?

Why can’t I find a particular file?

What the badges mean

What cloud placeholders are about

What to do about a signature mismatch

Found by filename, no badge, but the content still isn’t searchable

Not found even by filename

Can I add files from a NAS?

Why can’t I search what was said inside my audio and video files?

How do I keep a conversation focused on a single document?

How many files can a cloud library hold? Is it based on file size?

Can I use the AI features on a machine without internet access (offline / intranet)?

What should I do if the Linux AppImage will not start?

Is there a Linux server or Docker version?

How do I use the CLI to access Linkly AI on another computer?

Where do I find the settings I need?

Can I change where the Linkly AI Space is?

How do I move app data to another disk?

Steps

Why is there still a folder at the old location after the move?

How do I get Linkly AI’s application runtime logs?

Option 1: Open from inside the app (recommended)

Option 2: Open the app data directory manually

What if the app crashes during startup?