Bootstrap skill evaluation lab repo
os-eval-lab-setupskillsetup L2★3
richfrem/agent-plugins-skills ↗What it does
Bootstrap skill evaluation lab repository
Best for
Bootstrapping isolated test repos with consistent eval harness before autoresearch iterations.
Inputs
- · lab-repo-path
- · skill-path
- · phase-number
- · optimization-metric
Outputs
- · lab-repo-with-eval-engine
- · eval-instructions.md
Requires
- · Task/Subagent
- · git
- · Write
- · Bash
- · Read
Preconditions
- · Valid git repository initialized
- · Evaluation harness installed
Failure modes
- · Lab and master versions have diverged
Trust signals
- · Multi-phase execution with gates
- · Empirical evaluation scoring