Fix: coerce numeric arguments in ToolManager

[BhoomiAgrawal12]

Pre-PR Checklist

• [T] This PR is a bug fix, not a new feature or enhancement.

Summary

Fixes a bug where ToolManager passed numeric arguments as strings, causing tools annotated with int/float to fail. This impacted tool execution for LLMs that return numbers as JSON strings (e.g., Ollama).

ToolManager promises automatic type coercion in its docstring, but _process_tool_call() only filtered args by name. When an LLM returned "35.0" (string) for a float parameter, the tool received a string and raised an error. Expected behavior is to coerce to numeric types before invocation.

Bug / Issue

Fixes #276

Implementation

• Added coercion in _process_tool_call() for float and int parameters.
• Added two regression tests in tests/test_tools/test_tool_manager.py:
  • string → float ("35.0" -> 35.0)
  • string → int ("5" -> 5)

Testing

• pytest tests/test_tools/test_tool_manager.py -v

Additional Notes

• Coercion is intentionally limited to int and float to avoid unintended conversions.
• If coercion fails (invalid value), the original input is passed through so tools still surface meaningful errors.

[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: 3a88ac8d-02b2-4626-953a-4fb4d6e89cdb

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

---

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.}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 91.31%. Comparing base (2dfd129) to head (43f035f).

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #277      +/-   ##
==========================================
+ Coverage   91.20%   91.31%   +0.10%     
==========================================
  Files          19       19              
  Lines        1592     1612      +20     
==========================================
+ Hits         1452     1472      +20     
  Misses        140      140              

☔ 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.