Skip to content

GPT-5.2 not clearly prompted they shouldn't end conversation turn for tool preambles. #7900

@Nek-12

Description

@Nek-12

What version of Codex is running?

0.71.0

What subscription do you have?

Pro

Which model were you using?

5.2

What platform is your computer?

N/A

What issue are you seeing?

With GPT-5.2 it seems that you have enabled tool preambles or at least you prompted Codex to leave tool preambles. But you didn't prompt Codex clearly enough that it shouldn't end the conversation turn when it is leaving messages for the user. In the prompt the language is sometimes unclear and there are conflicting instructions. You need to update the system prompt to clearly show to Codex that it shouldn't end its conversation turn when it issues tool preambles. Right now Codex frequently exhibits the behavior of stopping before calling any tools.

One example:

Explored
  └ List core
    Search hook in core
    Search AGENTS.md
    Search StorePlugin in *Test*

─ Worked for 39s 

• Found PluginCallbacksTest.kt; opening it and the testing guide next.
  └ Read testing.md, PluginCallbacksTest.kt

─ Worked for 46s 


─ Worked for 2m 32s 

• There’s a TestPipelineContext helper in the test module; I’ll use it so we can actually invoke composed hooks too.

• Explored
  └ Read PluginTestScope.kt, TestPipelineContext.kt
    Search fun .*StorePlugin.*test in test
    Search PluginTestScope|StorePlugin\.test in test
    Read PluginTestDsl.kt

• Updated Plan
  └ ✔ Locate existing hooks tests
    ✔ Review testing guidelines
    □ Implement exhaustive composition test
    □ Run formatter and relevant tests

─ Worked for 3m 33s 

• Implementing a new Kotest in core/jvmTest that reflection-checks hook exhaustiveness and invokes each composed hook once.

// incorrectly stopped

› When you leave updates for the user - "me", you don't stop and end your conversation turn, you output text blocks before your tool calls without ending the conversation turn

// incorrectly stopped

• Got it, Nikita. I’ll keep progress updates as short preambles before tool calls, without yielding the turn.


› continue

Encountered 5+ times over 3 hours, decided to report. feel free to close if already fixed.

What steps can reproduce the bug?

N/A - model behavior

What is the expected behavior?

N/A - model behavior

Additional information

No response

Metadata

Metadata

Assignees

No one assigned

    Labels

    agentIssues related to the core agent loopbugSomething isn't working

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions