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 #3425] Custom AgentFlow - not sure if builder is able to do that #2211

New Issue
Closed
opened 2026-02-22 18:28:40 -05:00 by yindo · 1 comment
yindo commented 2026-02-22 18:28:40 -05:00
Owner
Copy Link
Copy Source

Originally created by @ruhlandm1 on GitHub (Mar 8, 2025).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/3425

Hi I´m trying to build a product research agent but not sure if I get it right. The idea is to do the following:

  1. Search for 5 suitable products referring to the user input (e.g. Laptop with aluminum body, i5 10th generation or newer, e.g.)
  2. Take those 5 products and look for reviews of them - choose the best two
  3. Scrap the internet for the best place to buy them and compare it to kleinanzeigen.de to get prices of used articles

In my mind I describe these three steps in detail and just combine three blocks of LLM Instruction to get something like this:

Image

Is that right in theory? What do I use as Input for the first Block?
Or should I simply use prompt engineering for that?

Originally created by @ruhlandm1 on GitHub (Mar 8, 2025). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/3425 Hi I´m trying to build a product research agent but not sure if I get it right. The idea is to do the following: 1. Search for 5 suitable products referring to the user input (e.g. Laptop with aluminum body, i5 10th generation or newer, e.g.) 2. Take those 5 products and look for reviews of them - choose the best two 3. Scrap the internet for the best place to buy them and compare it to kleinanzeigen.de to get prices of used articles In my mind I describe these three steps in detail and just combine three blocks of LLM Instruction to get something like this: ![Image](https://github.com/user-attachments/assets/dff23811-6260-4eab-a750-daad91b433e8) Is that right in theory? What do I use as Input for the first Block? Or should I simply use prompt engineering for that?
yindo closed this issue 2026-02-22 18:28:40 -05:00
yindo commented 2026-02-22 18:28:40 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Mar 9, 2025):

Ideally, you would use the input prompt to call a single tool/flow and that flow do a single series of steps correctly. LLMs are able to chain flows together so doing the same task 5 times, like for research, should just be using one tool - five times.

@timothycarambat commented on GitHub (Mar 9, 2025): Ideally, you would use the _input prompt_ to call a single tool/flow and that flow do a single series of steps correctly. LLMs are able to chain flows together so doing the same task 5 times, like for research, should just be using one tool - five times.
yindo changed title from Custom AgentFlow - not sure if builder is able to do that to [GH-ISSUE #3425] Custom AgentFlow - not sure if builder is able to do that 2026-06-05 14:45:11 -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
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#2211
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?