fix(llm_agent): initialize self.recorder so @record_model can attach it

[yashhzd]

Summary

LLMAgent.__init__ never initialized self.recorder, so the hasattr(agent, "recorder") check in _attach_recorder_to_agents() always returned False. This meant the @record_model decorator silently never attached the SimulationRecorder to any LLMAgent instance.

Changes

• mesa_llm/llm_agent.py: Add self.recorder: SimulationRecorder | None = None in __init__, with a TYPE_CHECKING import to avoid circular imports.
• tests/test_llm_agent.py: Add unit test verifying hasattr(agent, "recorder") returns True and defaults to None.
• tests/test_recording/test_record_model.py: Add integration test verifying @record_model successfully propagates the recorder to LLMAgent instances created during model init.

Test plan

• pytest tests/test_llm_agent.py::test_llm_agent_has_recorder_attribute -xvs passes
• pytest tests/test_recording/test_record_model.py::TestRecordModelDecorator::test_recorder_attached_to_llm_agents -xvs passes
• Full test suite (284 tests) passes with no regressions
• ruff check and all pre-commit hooks pass

Closes #218

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: b1de6a1f-8d39-4114-83ec-1dfa4bb49171

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment

📝 Coding Plan

• Generate coding plan for human review comments

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

Tip

CodeRabbit can use your project's `ruff` configuration to improve the quality of Python code reviews.

Add a Ruff configuration file to your project to customize how CodeRabbit runs ruff.

[codecov]

Codecov Report

❌ Patch coverage is 80.00000% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 90.64%. Comparing base (8e47d7b) to head (1267dce).

Files with missing lines	Patch %	Lines
mesa_llm/llm_agent.py	80.00%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #232      +/-   ##
==========================================
- Coverage   90.67%   90.64%   -0.04%     
==========================================
  Files          19       19              
  Lines        1555     1560       +5     
==========================================
+ Hits         1410     1414       +4     
- Misses        145      146       +1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[IlamaranMagesh]

Thanks for opening a PR. Though this is clearly a bug, I'm still waiting for some clarifications from the maintainers (#219) that I mentioned in the issue #218

[BhoomiAgrawal12]

hey @yashhzd! initializing self.recorder makes sense and fixes the issue of not attaching the recorder.

I had a small question regarding this part:

if TYPE_CHECKING:
    from mesa_llm.recording.simulation_recorder import SimulationRecorder

Since the file already uses future import annotations, the type hint should work without needing this import at all.

Given that SimulationRecorder isn’t directly used at runtime in this file, do we still need the TYPE_CHECKING block? Removing it might simplify things and also resolve the Codecov uncovered line.

[yashhzd]

Hey @BhoomiAgrawal12, great observation! I looked into this.

While from __future__ import annotations does make all annotations into strings at runtime, ruff's F821 rule still requires the name to be importable at static analysis time. Without the TYPE_CHECKING block, ruff reports Undefined name 'SimulationRecorder'. And using a string literal like "SimulationRecorder | None" triggers ruff's UP037 (remove quotes from type annotation) which then circles back to F821.

So both __future__ annotations and TYPE_CHECKING are actually needed together here:

• TYPE_CHECKING → satisfies ruff's static analysis (name is defined for type checkers)
• __future__ annotations → ensures the annotation is a string at runtime (prevents NameError since the import doesn't execute)

Regarding the Codecov patch failure — if TYPE_CHECKING: blocks are always False at runtime, which is a well-known coverage limitation. The project's [tool.coverage.report].exclude_lines in pyproject.toml doesn't currently include "if TYPE_CHECKING:". Adding it would be a project-wide config fix (not scoped to this PR). Happy to open a separate small PR for that if maintainers agree.