{"data":{"kind":"file","path":"README.md","version_id":"saa46vrrc6fo7xn93bl9n4lg","entry":{"name":"README.md","path":"README.md","is_directory":false,"size":1087,"modified_at":"2025-10-28T23:01:57.815000","content_hash":"5765c50eec862bbedd42bc6696e71e83d6b96b229246e2bc0d70000f0a552f7b"},"entries":[],"content":"# password-generator\n\n### Overview\n- **Environment ID**: `password-generator`\n- **Short description**: A password generator game where LLMs must create passwords satisfying randomly sampled rules.\n- **Tags**: password, generation, rules, single-turn\n\n### Datasets\n- **Primary dataset(s)**: None (prompts generated on-the-fly)\n- **Source links**: N/A\n- **Split sizes**: N/A\n\n### Task\n- **Type**: single-turn\n- **Parser**: None\n- **Rubric overview**: reward = sum(respected_rules) / len(rules)\n\n### Quickstart\nRun an evaluation with default settings:\n\n```bash\nuv run vf-eval password-generator\n```\n\nConfigure model and sampling:\n\n```bash\nuv run vf-eval password-generator   -m gpt-4.1-mini   -n 20 -r 3 -t 1024 -T 0.7   -a '{\"key\": \"value\"}'  # env-specific args as JSON\n```\n\nNotes:\n- Use `-a` / `--env-args` to pass environment-specific configuration as a JSON object.\n\n### Environment Arguments\nNo additional arguments supported.\n\n### Metrics\n| Metric | Meaning |\n| ------ | ------- |\n| `reward` | Reward is the number of rules that are respected divided by the total number of rules. |\n","encoding":"utf-8","truncated":false,"total_bytes":1087},"status":null}