Mintplex-Labs/anything-llm
1
0
Fork 0
mirror of https://github.com/Mintplex-Labs/anything-llm.git synced 2026-07-21 01:25:28 -04:00
Code Issues 328 Packages Projects Releases 10 Wiki Activity

[GH-ISSUE #5020] [BUG]: Purported features aren't real #3150

New Issue
Closed
opened 2026-02-22 18:32:49 -05:00 by yindo · 2 comments
yindo commented 2026-02-22 18:32:49 -05:00
Owner
Copy Link
Copy Source

Originally created by @lee-b on GitHub (Feb 18, 2026).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/5020

How are you running AnythingLLM?

AnythingLLM desktop app

What happened?

Marketing blurb says:

By default, the Desktop Assistant is able to chat with any open application using the full context of any application (Chrome, Slack, VS Code, etc.).

In a single keystroke, you can open the Desktop Assistant and start chatting, creating agents, and more with full context from any open applications.

But the reality is:

Do you know what’s on screen right now?

No, I cannot see what’s on your screen. I don’t have access to your display, your files, or any visual information about your computer. I can only see and respond to what you type to me in this conversation. If you’d like to share something with me, please paste or type it directly into the chat.

So how is this even meant to work then?

I work through text-based conversation:
You tell me things — I can only work with what you type to me directly in this chat. If you want me to help with code, you paste it here. If you have a question, you ask it here.
I respond to text — I read what you write and respond with text. I don’t have eyes, I can’t access your files, and I can’t see your screen.
Context comes from you — The only “context” I have is whatever you explicitly share in our conversation. The system prompt is just telling me to answer your follow-up question based on what you’ve told me in this chat.
So if you want help with something specific — like debugging code, explaining a concept, writing something, or answering a question — just type it out in plain text and I’ll do my best to help.
What would you like to work on?

Are there known steps to reproduce?

No response

Originally created by @lee-b on GitHub (Feb 18, 2026). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/5020 ### How are you running AnythingLLM? AnythingLLM desktop app ### What happened? Marketing blurb says: > By default, the Desktop Assistant is able to chat with any open application using the full context of any application (Chrome, Slack, VS Code, etc.). > In a single keystroke, you can open the Desktop Assistant and start chatting, creating agents, and more **with full context from any open applications.** But the reality is: > Do you know what’s on screen right now? > > No, I cannot see what’s on your screen. I don’t have access to your display, your files, or any visual information about your computer. I can only see and respond to what you type to me in this conversation. If you’d like to share something with me, please paste or type it directly into the chat. > So how is this even meant to work then? > > I work through text-based conversation: > > You tell me things — I can only work with what you type to me directly in this chat. If you want me to help with code, you paste it here. If you have a question, you ask it here. > > I respond to text — I read what you write and respond with text. I don’t have eyes, I can’t access your files, and I can’t see your screen. > > Context comes from you — The only “context” I have is whatever you explicitly share in our conversation. The system prompt is just telling me to answer your follow-up question based on what you’ve told me in this chat. > > So if you want help with something specific — like debugging code, explaining a concept, writing something, or answering a question — just type it out in plain text and I’ll do my best to help. > > What would you like to work on? ### Are there known steps to reproduce? _No response_
yindo added the possible bug label 2026-02-22 18:32:49 -05:00
yindo closed this issue 2026-02-22 18:32:49 -05:00
yindo commented 2026-02-22 18:32:49 -05:00
Author
Owner
Copy Link
Copy Source

@shatfield4 commented on GitHub (Feb 18, 2026):

The desktop assistant cannot view your screen in real time by just asking. It leverages screenshots of the applications on your machine to be passed as context to the LLM.

When you start a new thread, we are looking for the last used application you had open and there is a button to add that application as a screenshot to the LLM in one click by clicking the "Add to chat as a screenshot" button.

You can also do this by clicking the + icon in the chat input box and it will allow you to add your context from any app you have open on your machine. We also allow you to select entire displays, capture certain areas of your screen, or chat with your meetings from the meeting assistant in AnythingLLM. This allows you to give screenshots of whatever you have open on your machine as context to the LLM.

@shatfield4 commented on GitHub (Feb 18, 2026): The desktop assistant cannot view your screen in real time by just asking. It leverages screenshots of the applications on your machine to be passed as context to the LLM. When you start a new thread, we are looking for the last used application you had open and there is a button to add that application as a screenshot to the LLM in one click by clicking the "Add to chat as a screenshot" button. You can also do this by clicking the + icon in the chat input box and it will allow you to add your context from any app you have open on your machine. We also allow you to select entire displays, capture certain areas of your screen, or chat with your meetings from the meeting assistant in AnythingLLM. This allows you to give screenshots of whatever you have open on your machine as context to the LLM.
yindo commented 2026-02-22 18:32:49 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Feb 18, 2026):

The video of the demo shows exactly how the tool works - and its exactly as described. We wont read your screen with permission since that is a security violation in our opinion.
https://docs.anythingllm.com/desktop-assistant/introduction

@timothycarambat commented on GitHub (Feb 18, 2026): The video of the demo shows _exactly_ how the tool works - and its exactly as described. We wont read your screen with permission since that is a security violation in our opinion. https://docs.anythingllm.com/desktop-assistant/introduction
yindo changed title from [BUG]: Purported features aren't real to [GH-ISSUE #5020] [BUG]: Purported features aren't real 2026-06-05 14:50:34 -04:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
Desktop
Docker
Integration Request
Integration Request
OS: Linux
OS: Mobile
OS: Windows
UI/UX
blocked
bug
bug
core-team-only
documentation
duplicate
embed-widget
enhancement
feature request
github_actions
good first issue
investigating
needs info / can't replicate
possible bug
pull-request
Mirrored from GitHub Pull Request
 question
stage: specifications
wontfix
No Label possible bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Mintplex-Labs/anything-llm#3150
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?