langchain-ai/docs
1
0
Fork 0
mirror of https://github.com/langchain-ai/docs.git synced 2026-07-22 12:25:55 -04:00
Code Issues 490 Packages Projects Releases Wiki Activity

[GH-ISSUE #1435] [langchain]: The “after agent guardrails” example doesn’t work properly #206

New Issue
Closed
opened 2026-02-17 17:19:24 -05:00 by yindo · 0 comments
yindo commented 2026-02-17 17:19:24 -05:00
Owner
Copy Link
Copy Source

Originally created by @mcavdar on GitHub (Nov 15, 2025).
Original GitHub issue: https://github.com/langchain-ai/docs/issues/1435

Type of issue

issue / bug

Language

Python

Description

Hi, I'm running the “After Agent Guardrails” example from the LangChain documentation using the Class Syntax version.

When the judge LLM returns “UNSAFE”, the agent enters an infinite loop.
I believe the code needs an update, specifically in this section of the example.

        if "UNSAFE" in result.content:
            return {
                "messages": [{
                    "role": "assistant",
                    "content": "I cannot provide that response. Please rephrase your request."
                }],
                "jump_to": "end"
            }

ChatGPT says

Your after_agent middleware is triggering AGAIN AND AGAIN in a loop

This happens because:

after_agent runs every time the agent produces a message.

Your middleware adds a new assistant message when unsafe.

That added assistant message makes the graph think another agent turn is required.

The agent runs again → produces new message → middleware runs again → loops forever.

Originally created by @mcavdar on GitHub (Nov 15, 2025). Original GitHub issue: https://github.com/langchain-ai/docs/issues/1435 ### Type of issue issue / bug ### Language Python ### Description Hi, I'm running the “After Agent Guardrails” [example](https://docs.langchain.com/oss/python/langchain/guardrails#after-agent-guardrails) from the LangChain documentation using the Class Syntax version. When the judge LLM returns “UNSAFE”, the agent enters an **infinite loop**. I believe the code needs an update, specifically in this section of the example. ``` if "UNSAFE" in result.content: return { "messages": [{ "role": "assistant", "content": "I cannot provide that response. Please rephrase your request." }], "jump_to": "end" } ``` ChatGPT says > Your after_agent middleware is triggering AGAIN AND AGAIN in a loop > > This happens because: > > after_agent runs every time the agent produces a message. > > Your middleware adds a new assistant message when unsafe. > > That added assistant message makes the graph think another agent turn is required. > > The agent runs again → produces new message → middleware runs again → loops forever.
yindo added the langchain label 2026-02-17 17:19:24 -05:00
yindo closed this issue 2026-02-17 17:19:24 -05:00
yindo changed title from [langchain]: The “after agent guardrails” example doesn’t work properly to [GH-ISSUE #1435] [langchain]: The “after agent guardrails” example doesn’t work properly 2026-06-05 17:25:38 -04:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
bug
deepagents
docs-infra
external
external
help wanted
integration
internal
internal
javascript
js/ts
langchain
langgraph
langsmith
needs-title
pull-request
Mirrored from GitHub Pull Request
 python
reference
No Label langchain
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/docs#206
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?