Mintplex-Labs/anything-llm
1
0
Fork 0
mirror of https://github.com/Mintplex-Labs/anything-llm.git synced 2026-07-19 14:13:35 -04:00
Code Issues 328 Packages Projects Releases 10 Wiki Activity

[GH-ISSUE #5576] [NOT-BUG]: listing of documents behaviour ? #5174

New Issue
Closed
opened 2026-06-05 14:52:27 -04:00 by yindo · 3 comments
yindo commented 2026-06-05 14:52:27 -04:00
Owner
Copy Link
Copy Source

Originally created by @ohromeo on GitHub (May 4, 2026).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/5576

How are you running AnythingLLM?

Docker (remote machine)

What happened?

v1.12.0

files in the workspace:

Image

Question Posed:
give me a brief overview of the documents in my workspace

Response:
(details excluded...)

Based on the content provided, there are two documents in your workspace:

  1. Conflict Resolution.pdf
  2. Accessibility Policy_Final.pdf

as you'll notice Dress Code.en-US.pdf is missing... this was a freshly uploaded document and to queries were made to it...

Question Posed:
/sum Dress Code.en-US.pdf

Response:
response received,

Then ran the initial request again

Question Posed:
give me a brief overview of the documents in my workspace

Response:
(details excluded...)
1 | Conflict Resolution.pdf | HR Policy | February 2024 | February 2027
2 | Accessibility Policy_Final.pdf | HR Policy | Not specified | Every 3 years
3 | Dress Code.en-US.pdf | HR Policy | February 2024 | February 2027

it now returned the 3rd document... is this by design/behaviour?, that if the document hasn't been accessed, it will not be in the "knowledge base" and thus invisible even though it is embedded in the workspace ?

notes:
this is through the api calls from my "front-end", same behaviour with ALLM interface

thanks,
JS

Are there known steps to reproduce?

upload a document to your workspace -- don;t perform any functions on it.
ask the agent to tell you about your documents in the workspace

LLM Provider & Model (if applicable)

generic-openai - deepseek-chat

Embedder Provider & Model (if applicable)

No response

Originally created by @ohromeo on GitHub (May 4, 2026). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/5576 ### How are you running AnythingLLM? Docker (remote machine) ### What happened? v1.12.0 files in the workspace: <img width="381" height="173" alt="Image" src="https://github.com/user-attachments/assets/ff58c260-ab5d-4a45-b522-dcc622d505ec" /> Question Posed: give me a brief overview of the documents in my workspace Response: (details excluded...) Based on the content provided, there are **two documents** in your workspace: 1. **Conflict Resolution.pdf** 2. **Accessibility Policy_Final.pdf** as you'll notice Dress Code.en-US.pdf is missing... this was a freshly uploaded document and to queries were made to it... Question Posed: /sum Dress Code.en-US.pdf Response: response received, Then ran the initial request again Question Posed: give me a brief overview of the documents in my workspace Response: (details excluded...) 1 | Conflict Resolution.pdf | HR Policy | February 2024 | February 2027 2 | Accessibility Policy_Final.pdf | HR Policy | Not specified | Every 3 years 3 | Dress Code.en-US.pdf | HR Policy | February 2024 | February 2027 it now returned the 3rd document... is this by design/behaviour?, that if the document hasn't been accessed, it will not be in the "knowledge base" and thus invisible even though it is embedded in the workspace ? notes: this is through the api calls from my "front-end", same behaviour with ALLM interface thanks, JS ### Are there known steps to reproduce? upload a document to your workspace -- don;t perform any functions on it. ask the agent to tell you about your documents in the workspace ### LLM Provider & Model (if applicable) generic-openai - deepseek-chat ### Embedder Provider & Model (if applicable) _No response_
yindo added the possible bugneeds info / can't replicate labels 2026-06-05 14:52:28 -04:00
yindo closed this issue 2026-06-05 14:52:28 -04:00
yindo commented 2026-06-05 14:52:28 -04:00
Author
Owner
Copy Link
Copy Source

@angelplusultra commented on GitHub (May 4, 2026):

Are you using @agent for your prompts or are your prompts exactly as you laid out here (e.g "give me a brief overview of the documents in my workspace")

<!-- gh-comment-id:4375075381 --> @angelplusultra commented on GitHub (May 4, 2026): Are you using `@agent` for your prompts or are your prompts exactly as you laid out here (e.g "give me a brief overview of the documents in my workspace")
yindo commented 2026-06-05 14:52:28 -04:00
Author
Owner
Copy Link
Copy Source

@ohromeo commented on GitHub (May 5, 2026):

Hi,

Same as posted, chat mode was on auto.

I can redo the scenario if you need any additional information. I have upgraded my linux docker to the latest version as well.

thanks!

From: Marcello Fitton @.>
Sent: Monday, May 4, 2026 6:49 PM
To: Mintplex-Labs/anything-llm @.>
Cc: ohromeo @.>; Author @.>
Subject: Re: [Mintplex-Labs/anything-llm] [NOT-BUG]: listing of documents behaviour ? (Issue #5576)

[https://avatars.githubusercontent.com/u/106866560?s=20&v=4]angelplusultra left a comment (Mintplex-Labs/anything-llm#5576)https://github.com/Mintplex-Labs/anything-llm/issues/5576#issuecomment-4375075381

Are you using @agent for your prompts or are your prompts exactly as you laid out here (e.g give me a brief overview of the documents in my workspace)

—
Reply to this email directly, view it on GitHubhttps://github.com/Mintplex-Labs/anything-llm/issues/5576#issuecomment-4375075381, or unsubscribehttps://github.com/notifications/unsubscribe-auth/BDJHFKVJGXZWQW4PB3UBKLL4ZENARAVCNFSM6AAAAACYQQXCFSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHM2DGNZVGA3TKMZYGE.
Triage notifications on the go with GitHub Mobile for iOShttps://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Androidhttps://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.
You are receiving this because you authored the thread.Message ID: @.***>

<!-- gh-comment-id:4378542800 --> @ohromeo commented on GitHub (May 5, 2026): Hi, Same as posted, chat mode was on auto. I can redo the scenario if you need any additional information. I have upgraded my linux docker to the latest version as well. thanks! ________________________________ From: Marcello Fitton ***@***.***> Sent: Monday, May 4, 2026 6:49 PM To: Mintplex-Labs/anything-llm ***@***.***> Cc: ohromeo ***@***.***>; Author ***@***.***> Subject: Re: [Mintplex-Labs/anything-llm] [NOT-BUG]: listing of documents behaviour ? (Issue #5576) [https://avatars.githubusercontent.com/u/106866560?s=20&v=4]angelplusultra left a comment (Mintplex-Labs/anything-llm#5576)<https://github.com/Mintplex-Labs/anything-llm/issues/5576#issuecomment-4375075381> Are you using @agent for your prompts or are your prompts exactly as you laid out here (e.g give me a brief overview of the documents in my workspace) — Reply to this email directly, view it on GitHub<https://github.com/Mintplex-Labs/anything-llm/issues/5576#issuecomment-4375075381>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/BDJHFKVJGXZWQW4PB3UBKLL4ZENARAVCNFSM6AAAAACYQQXCFSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHM2DGNZVGA3TKMZYGE>. Triage notifications on the go with GitHub Mobile for iOS<https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675> or Android<https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>. You are receiving this because you authored the thread.Message ID: ***@***.***>
yindo commented 2026-06-05 14:52:28 -04:00
Author
Owner
Copy Link
Copy Source

@angelplusultra commented on GitHub (May 5, 2026):

If I’m understanding your use case correctly, this looks like a common misunderstanding of how workspace documents function in AnythingLLM.

Prompts like “Summarize the content of file X” or “List all files in this workspace” won’t work in a regular chat (i.e., without @agent). In standard chats, document interaction is powered by a basic RAG pipeline: it retrieves semantically relevant chunks from documents in your workspace and injects them into the model’s context. Because of this, you’re not interacting with the documents as discrete objects—you’re only interacting with their content.

For example:

  • If you have a document like rome.pdf in your workspace:
    • ✅ “Tell me about the construction of the Colosseum” → works (content-based query)
    • ❌ “Summarize rome.pdf” or “List all files in the workspace” → won’t work (document-level queries)

This limitation is inherent to how RAG pipelines operate.

You’ll need to use agent-based interactions instead of regular chat.

  1. In AnythingLLM, go to Settings → Agent Skills
  2. Ensure “View & Summarize Documents” is enabled
  3. Use agent prompts by prepending @agent, for example:
    @agent give me a brief overview of the documents in my workspace

When using agent mode, the system can call tools like document-summarizer, which do have awareness of the workspace as a collection (including listing files and summarizing them).

You should see the agent explicitly invoke this tool when running your prompt. Heres an example of this using the AnythingLLM CLI which interfaces with the API.

Image
<!-- gh-comment-id:4381771204 --> @angelplusultra commented on GitHub (May 5, 2026): If I’m understanding your use case correctly, this looks like a common misunderstanding of how workspace documents function in AnythingLLM. Prompts like “Summarize the content of file X” or “List all files in this workspace” won’t work in a regular chat (i.e., without @agent). In standard chats, document interaction is powered by a basic RAG pipeline: it retrieves semantically relevant chunks from documents in your workspace and injects them into the model’s context. Because of this, you’re not interacting with the documents as discrete objects—you’re only interacting with their **content**. For example: * If you have a document like `rome.pdf` in your workspace: * ✅ “Tell me about the construction of the Colosseum” → works (content-based query) * ❌ “Summarize rome.pdf” or “List all files in the workspace” → won’t work (document-level queries) This limitation is inherent to how RAG pipelines operate. You’ll need to use agent-based interactions instead of regular chat. 1. In AnythingLLM, go to Settings → Agent Skills 2. Ensure “View & Summarize Documents” is enabled 3. Use agent prompts by prepending @agent, for example: @agent give me a brief overview of the documents in my workspace When using agent mode, the system can call tools like `document-summarizer`, which do have awareness of the workspace as a collection (including listing files and summarizing them). You should see the agent explicitly invoke this tool when running your prompt. Heres an example of this using the AnythingLLM CLI which interfaces with the API. <img width="2872" height="482" alt="Image" src="https://github.com/user-attachments/assets/1fe3af93-7a61-4d30-be89-86aa8ac01699" />
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 needs info / can't replicate 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#5174
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?