{"data":{"kind":"file","path":"README.md","version_id":"kgv9iepwqmy66djnpny0dech","entry":{"name":"README.md","path":"README.md","is_directory":false,"size":1013,"modified_at":"2025-12-24T20:24:30.286000","content_hash":"845200b51706fdcb628d2c725500f1815fbb2a9e5694ab1bda7f54790c03f8ce"},"entries":[],"content":"# Code Repair Environment\n\nA verifiers-compatible RL environment for training LLMs to fix buggy Python code.\n\n## Installation\n`\bash\npip install -e .\n`\n\n## Usage\n`python\nfrom code_repair import load_environment\n\n# Create environment with 100 buggy code examples\nenv = load_environment(num_examples=100)\n\n# View a sample\nprint(env.dataset[0])\n`\n\n## Bug Types\n\n- wrong_operator (+ instead of *)\n- off_by_one (index errors)\n- wrong_base_case (recursion errors)\n- missing_guard (no null checks)\n- type_error (string/int confusion)\n- type_conversion (missing int()/str())\n- And more...\n\n## With OpenAI API\n`python\nfrom openai import OpenAI\nfrom code_repair import load_environment\n\nenv = load_environment(num_examples=10)\nclient = OpenAI()\n\nresults = env.evaluate(\n    client=client,\n    model=\"gpt-4.1-mini\",\n    num_examples=10\n)\n`\n\n## Prime Intellect Integration\n\nThis environment is compatible with the Prime Intellect verifiers framework and can be published to their Environments Hub.\n\n## License\n\nApache 2.0\r\n","encoding":"utf-8","truncated":false,"total_bytes":1013},"status":null}