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: System Prompt in Create ReAct agent is un effective #366

New Issue
Closed
opened 2026-02-20 17:38:42 -05:00 by yindo · 1 comment
yindo commented 2026-02-20 17:38:42 -05:00
Owner
Copy Link
Copy Source

Originally created by @DucretJe on GitHub (Dec 19, 2024).

Issue with current documentation:

In the nodes and edge definition from How to create a ReAct agent from scratch, we declare a system_prompt variable and pass it to the model.invoke along the state.

This has in fact no effect, I replaced the prompt with:

    system_prompt = SystemMessage(
        "Speak in Italian"
    )

We get an english answer.

I also noticed that we had some warnings with this

response = model.invoke([system_prompt] + state["messages"], config)

Operator "+" not supported for types "list[SystemMessage]" and "Sequence[BaseMessage]"

Idea or request for content:

I managed to make it work with this:

    system_prompt = SystemMessage(
        "Speak in Italian"
    )
    state["messages"].insert(0, system_prompt)
    response = model.invoke(state["messages"], config)

Doing so I'm assured to have one first message containing the system prompt.

Note

If the same node is called several times (which is the case when a tool redirect to this node), the system prompt is injected multiple times... So I guess we can introduce a test checking that the first message is a System prompt.
We can imagine something like:

if not isinstance(state["messages"][0], SystemMessage):
  state["messages"].insert(0, system_prompt)
Originally created by @DucretJe on GitHub (Dec 19, 2024). ### Issue with current documentation: In the [nodes and edge definition](https://langchain-ai.github.io/langgraph/how-tos/react-agent-from-scratch/) from `How to create a ReAct agent from scratch`, we declare a `system_prompt` variable and pass it to the `model.invoke` along the state. This has in fact no effect, I replaced the prompt with: ```python system_prompt = SystemMessage( "Speak in Italian" ) ``` We get an english answer. I also noticed that we had some warnings with this ```python response = model.invoke([system_prompt] + state["messages"], config) ``` > Operator "+" not supported for types "list[SystemMessage]" and "Sequence[BaseMessage]" ### Idea or request for content: I managed to make it work with this: ```python system_prompt = SystemMessage( "Speak in Italian" ) state["messages"].insert(0, system_prompt) response = model.invoke(state["messages"], config) ``` Doing so I'm assured to have one first message containing the system prompt. > [!NOTE] > If the same node is called several times (which is the case when a tool redirect to this node), the system prompt is injected multiple times... So I guess we can introduce a test checking that the first message is a System prompt. > We can imagine something like: > ```python > if not isinstance(state["messages"][0], SystemMessage): > state["messages"].insert(0, system_prompt) > ```
yindo closed this issue 2026-02-20 17:38:42 -05:00
yindo commented 2026-02-20 17:38:46 -05:00
Author
Owner
Copy Link
Copy Source

@vbarda commented on GitHub (Dec 19, 2024):

@DucretJe you shouldn't be modifying the state object manually, it should only be modified by the updates from the nodes (i.e. return from the node function) - this is also why you're seeing the issue of the message getting injected multiple times when you modify the state manually

this is the correct way of doing it - system message is dynamically added ONLY when the model is called, but NOT added to the graph state

system_prompt = SystemMessage("Speak in Italian")
response = model.invoke([system_prompt] + state["messages"], config)

hope this helps. since this is not an issue with the documentation / langgraph, will close it, but feel free to open a discussion or another issue

@vbarda commented on GitHub (Dec 19, 2024): @DucretJe you shouldn't be modifying the state object manually, it should only be modified by the updates from the nodes (i.e. return from the node function) - this is also why you're seeing the issue of the message getting injected multiple times when you modify the state manually this is the correct way of doing it - system message is dynamically added ONLY when the model is called, but NOT added to the graph state ```python system_prompt = SystemMessage("Speak in Italian") response = model.invoke([system_prompt] + state["messages"], config) ``` hope this helps. since this is not an issue with the documentation / langgraph, will close it, but feel free to open a discussion or another issue
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#366
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?