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 #3922] [FEAT]: Expose all tools, agents, llm calls, user database tables in the handler.js #2499

New Issue
Closed
opened 2026-02-22 18:29:57 -05:00 by yindo · 1 comment
yindo commented 2026-02-22 18:29:57 -05:00
Owner
Copy Link
Copy Source

Originally created by @uninstallit on GitHub (May 30, 2025).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/3922

What would you like to see?

Perhaps, there are limitations that I am not aware of - but when I build a custom agent - I would like the ability to have access to the whole ecosystem within an agent.

For example, lets say I want to handle a different chunking strategy - I can create one or multiple agents that can chunk, embed, and save to database tables that pertain to the authenticated user.

I think it would also be nice for an agent to be able to call another agent from within the handler - after all its just a function. Following the example above - I can have a custom agent that performs rag multiple times on the different chunking strategies - and its invoking the chunking agents from before.

The ability to have more access to the ecosystem also pertains to the agent flows where there is no agent/custom agent block, database, tool, etc.

Originally created by @uninstallit on GitHub (May 30, 2025). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/3922 ### What would you like to see? Perhaps, there are limitations that I am not aware of - but when I build a custom agent - I would like the ability to have access to the whole ecosystem within an agent. For example, lets say I want to handle a different chunking strategy - I can create one or multiple agents that can chunk, embed, and save to database tables that pertain to the authenticated user. I think it would also be nice for an agent to be able to call another agent from within the handler - after all its just a function. Following the example above - I can have a custom agent that performs rag multiple times on the different chunking strategies - and its invoking the chunking agents from before. The ability to have more access to the ecosystem also pertains to the agent flows where there is no agent/custom agent block, database, tool, etc.
yindo added the enhancementfeature request labels 2026-02-22 18:29:57 -05:00
yindo closed this issue 2026-02-22 18:29:57 -05:00
yindo commented 2026-02-22 18:29:57 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Jun 3, 2025):

Since it is all imported code, you technically can bundle infinite functions into that handler.js file and import them at the top. This includes other agents as well - since you are calling them in the function itself the LLM does not need to explictly call them.

Ideally, an agent does one "task" and it is up to the LLM to chain them together for any given query. But if you want to invoke functions from other agents (or even other JS files), as long as the function is module.exported then you can import it and use it. There is no restriction there since it is all just evaluated as code inside the handler.

This is intentional, so that you really can do whatever you want in an agent skills.

Does that make sense? Do you have an example to share?

@timothycarambat commented on GitHub (Jun 3, 2025): Since it is all imported code, you technically can bundle infinite functions into that handler.js file and `import` them at the top. This includes other agents as well - since you are calling them in the function itself the LLM does not need to explictly call them. Ideally, an agent does one "task" and it is up to the LLM to chain them together for any given query. But if you want to invoke functions from other agents (or even other JS files), as long as the function is `module.export`ed then you can import it and use it. There is no restriction there since it is all just evaluated as code inside the handler. This is intentional, so that you really can do whatever you want in an agent skills. Does that make sense? Do you have an example to share?
yindo changed title from [FEAT]: Expose all tools, agents, llm calls, user database tables in the handler.js to [GH-ISSUE #3922] [FEAT]: Expose all tools, agents, llm calls, user database tables in the handler.js 2026-06-05 14:46:55 -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 enhancement feature request
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#2499
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?