langchain-ai/langgraph
1
0
Fork 0
mirror of https://github.com/langchain-ai/langgraph.git synced 2026-07-15 12:35:43 -04:00
Code Issues 314 Packages Projects Releases 17 Wiki Activity

DOC: Multi-agent Systems #301

New Issue
Closed
opened 2026-02-20 17:36:00 -05:00 by yindo · 1 comment
yindo commented 2026-02-20 17:36:00 -05:00
Owner
Copy Link
Copy Source

Originally created by @kellerkind84 on GitHub (Nov 5, 2024).

Issue with current documentation:

I think there is an issue with the "Supervisor (as tools)" graphic. It implies that the LLM picks from a number of tools. I don't think that is correct, since only agents pick from tools and everything that is not picked by an agent is a node and therefore requires an edge per node to be called. This is also correctly implemented in the example.

Idea or request for content:

Simply replace the "llm" with "agent".
You might as well add some tool-definition to the "Supervisor with tools" example for the tools, since it makes it sound like there is no tool definition necessary (which could also lead to no clear distinction between a node and a tool).

Originally created by @kellerkind84 on GitHub (Nov 5, 2024). ### Issue with current documentation: I think there is an issue with the "Supervisor (as tools)" graphic. It implies that the LLM picks from a number of tools. I don't think that is correct, since only agents pick from tools and everything that is not picked by an agent is a node and therefore requires an edge per node to be called. This is also correctly implemented in the example. ### Idea or request for content: Simply replace the "llm" with "agent". You might as well add some tool-definition to the "Supervisor with tools" example for the tools, since it makes it sound like there is no tool definition necessary (which could also lead to no clear distinction between a node and a tool).
yindo closed this issue 2026-02-20 17:36:00 -05:00
yindo commented 2026-02-20 17:36:00 -05:00
Author
Owner
Copy Link
Copy Source

@vbarda commented on GitHub (Nov 5, 2024):

@kellerkind84 which page are you referencing?

I think there is an issue with the "Supervisor (as tools)" graphic. It implies that the LLM picks from a number of tools. I don't think that is correct, since only agents pick from tools and everything that is not picked by an agent is a node and therefore requires an edge per node to be called. This is also correctly implemented in the example.

but the LLM does pick from a number of tools in this case -- the agent (supervisor) is simply a tool-calling LLM and the individual agents it's delegating to are wrapped as tools. i am not sure i understand your concern

You might as well add some tool-definition to the "Supervisor with tools" example for the tools, since it makes it sound like there is no tool definition necessary (which could also lead to no clear distinction between a node and a tool).

in the supervisor with tools example we use a prebuilt ReAct agent with functions as tools (e.g. agent_1 and agent_2 are tools in this case)

@vbarda commented on GitHub (Nov 5, 2024): @kellerkind84 which page are you referencing? > I think there is an issue with the "Supervisor (as tools)" graphic. It implies that the LLM picks from a number of tools. I don't think that is correct, since only agents pick from tools and everything that is not picked by an agent is a node and therefore requires an edge per node to be called. This is also correctly implemented in the example. but the LLM _does_ pick from a number of tools in this case -- the agent (supervisor) is simply a tool-calling LLM and the individual agents it's delegating to are wrapped as tools. i am not sure i understand your concern > You might as well add some tool-definition to the "Supervisor with tools" example for the tools, since it makes it sound like there is no tool definition necessary (which could also lead to no clear distinction between a node and a tool). in the [supervisor with tools example](https://langchain-ai.github.io/langgraph/concepts/multi_agent/#supervisor-tool-calling) we use a prebuilt ReAct agent with functions as tools (e.g. `agent_1` and `agent_2` are tools in this case)
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
agents
awaiting author revision
bug
change
chat models
checkpointer
ci
code quality
code quality
dependencies
documentation
drawing
duplicate
enhancement
good first issue
help wanted
invalid
maintainer
meta
missing information
not maintained by langchain
open-swe-auto
open-swe-max
open-swe-max-auto
pending
performance
pull-request
Mirrored from GitHub Pull Request
 question
requires design
sdk
server
stale
studio
tests
type checking
usage question
v0.5
v1.1
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: langchain-ai/langgraph#301
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?