domains / families / agent-tooling / run
run
separate
Run a single experiment iteration. Edit the target file, evaluate, keep or discard.
When to use
- Run a single experiment iteration
- Edit the target file
- Evaluate experiment changes
- Keep or discard experiment results
- Review experiment history
- Decide on a single change
- Commit and evaluate changes
- Report experiment results
When not to use
- Modifying the evaluator (evaluate.py)
- Changing multiple things at once
- Adding new dependencies
Skill page
Source