Run2 setup repo environment
[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
npx -y skills add cxcscmu/SkillLearnBench --skill run2_setup_repo_environmentAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
What its author says it does
Copied from the file, not written here
Aligns the local environment with the repository's requirements by checking for dependency files, installing build-essential tools for complex packages like DeepSpeed, and logging environment metadata.
SKILL.md
1.1 KB, as published. Nobody here has run it
- Identify Dependencies: Check the
/root/SimPOdirectory forenvironment.yml,requirements.txt, orpyproject.toml. - Align Python Version: If a specific version is required by
environment.yml, attempt to use that version. - Install Build Tools: Ensure the system has tools required for compiling C++/CUDA extensions (common in NLP repos for
deepspeedorflash-attn):apt-get update && apt-get install -y build-essential libaio-dev
- Install Packages:
- If
requirements.txtexists:pip install -r requirements.txt. - Ensure
torch,numpy, andtransformersare compatible with the versions mentioned in the repo.
- If
- Log Environment Info:
- Run
python -VV > /root/python_info.txt - Run
python -m pip freeze >> /root/python_info.txt - This ensures the exact state used for the loss computation is recorded for reproducibility.
- Run