The Agent Skills Directory

[COMMAND_EXECUTION]: The scripts/run_experiment.py utility executes shell commands specified in a local configuration file (.autoresearch/{experiment}/config.cfg). This is the intended functionality for running automated evaluations in an experiment loop.
Evidence: The script uses subprocess.run(config["evaluate_cmd"], shell=True) to execute the configured test command.
Safety Controls: The script implements several guardrails: it refuses to run unless the current Git branch matches an isolation pattern (autoresearch/{experiment}), and it halts if any files other than the experiment target have uncommitted changes in the working tree, preventing accidental side effects.
[NO_CODE]: The main SKILL.md file references an installation utility scripts/install-external-skills.py for connecting to external skill chains. This script is not included in the provided file set, which appears to be a documentation reference to a utility maintained in the parent repository.

loops