Bala

Abandoning as it requires significant changes in toolkit which will be taken separately. Will take it once the planned toolkit refactoring is complete.

Buddy-Build Successful: https://dev.azure.com/mariner-org/mariner/_build/results?buildId=663567&view=results

This error can be overcome by passing `null` for `top_p` as follows ``` sweagent run \ --env.repo.github_url https://github.com/SWE-agent/test-repo \ --problem_statement.github_url https://github.com/SWE-agent/test-repo/issues/1 \ --agent.model.name azure/o3 \ --agent.model.top_p null \ --agent.model.temperature 1...

@WordDealer In both the cases the call failed with unknown parameter `reasoning_effort`. It is understandable that you don't expect the error when you pass null as the value or when...

@WordDealer You're facing this error as `reasoning_effort` parameter is passed by the config file. As you didn't supply any config file in you command-line, it takes `config/default.yaml` as the default...

@WordDealer There is an open PR by @klieret to fix this issue - https://github.com/SWE-agent/SWE-agent/pull/1281/files It will get merged soon.

> I do not have an explanation for that; I can only share my reasoning so far: The token limits for a single call are [1M input and 66k output](https://cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/2-5-pro?utm_source=chatgpt.com)....

Here is your error message ``` error running container: from /usr/bin/crun creating container for [/bin/sh -c apt-get update && apt-get install -y wget gcc make zlib1g-dev libssl-dev && rm -rf...

It seems that `Check Markdown links` check failure is not related. It can be addressed in a separate PR.

@archana25-ms Please check the source-signature-check. And also, resolve the conflicts. @mfrw Re-Requesting review to verify above changes.