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 #1280] [FEAT]: @agent custom Large-action-model support with built-in training #801

New Issue
Closed
opened 2026-02-22 18:21:26 -05:00 by yindo · 0 comments
yindo commented 2026-02-22 18:21:26 -05:00
Owner
Copy Link
Copy Source

Originally created by @timothycarambat on GitHub (May 3, 2024).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/1280

Originally assigned to: @timothycarambat on GitHub.

What would you like to see?

Desktop only - this is not possible on docker.

The user should be able to define a custom LAM action with any LLM we support for @agent.

  • The user should be able to authenticate and "train" the LLM in what steps it should take to accomplish an action
    e.g: "Log in to my Shopify account and tell me all the pending orders I have today"

Training is simply opening a browser and the user using the browser as they normally would to accomplish said action. Nothing fancy needs to occur.

  • The user should be able to define and override steps in the flow and define steps as custom with a defined prompt/objective so that the flows are not too rigid to be useful.
    e.g: "Post a haiku to Twitter" => should be a different haiku each time

The user should be able to define these flows and when @agent calls them, we should record and or capture the UI for debugging purposes and log the runs for any of these LAM runs.

Originally created by @timothycarambat on GitHub (May 3, 2024). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/1280 Originally assigned to: @timothycarambat on GitHub. ### What would you like to see? **Desktop only - this is not possible on docker.** The user should be able to define a custom LAM action with any LLM we support for `@agent`. - The user should be able to authenticate and "train" the LLM in what steps it should take to accomplish an action e.g: "Log in to my Shopify account and tell me all the pending orders I have today" > Training is simply opening a browser and the user using the browser as they normally would to accomplish said action. Nothing fancy needs to occur. - The user should be able to define and override steps in the flow and define steps as `custom` with a defined prompt/objective so that the flows are not too rigid to be useful. e.g: "Post a haiku to Twitter" => should be a different haiku each time The user should be able to define these flows and when `@agent` calls them, we should record and or capture the UI for debugging purposes and log the runs for any of these LAM runs.
yindo added the enhancementcore-team-onlyDesktop labels 2026-02-22 18:21:27 -05:00
yindo closed this issue 2026-02-22 18:21:27 -05:00
yindo changed title from [FEAT]: `@agent` custom Large-action-model support with built-in training to [GH-ISSUE #1280] [FEAT]: `@agent` custom Large-action-model support with built-in training 2026-06-05 14:37:15 -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 Desktop core-team-only enhancement
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#801
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?