What problem does it solve?
Run existing ShinkaEvolve tasks with the shinka_run CLI from a task directory (evaluate.py + initial.<ext>). Use when an agent needs to launch async evolution runs quickly with required --results_dir, generation count, and strict namespaced keyword overrides.
Core Features & Use Cases
- Batch-run ShinkaEvolve tasks from a directory containing evaluate.py and initial.* to start controlled evolution experiments.
- Validate task readiness: ensures evaluate.py and initial.* exist before launching.
- CLI-driven orchestration: configure generations, results path, and overrides with explicit --set properties for reproducible runs.
- Suitable for agents or developers needing fast, asynchronous evolution runs with clear output directories.
Quick Start
Launch a batch evolution by pointing the CLI to your task directory and desired results directory, e.g., shinka_run --task-dir <task_dir> --results_dir <results_dir> --num_generations 40.