treant_forest.json

{
  "task_id": "treant_forest",
  "name": "Treant Forest",
  "category": "Combinatorial Optimization",
  "base_image": "python310",
  "platform": "linux/amd64",
  "internet": false,
  "cwd": "/home/workspace/treants_forest",
  "submit_paths": [
    "solution.py"
  ],
  "work": {
    "image_tag": "e05dfd5d2463",
    "specs_dir": "/home/workspace/treants_forest",
    "agent_query": "## Treant's Forest - Maze Obstruction Strategy (AHC054)\n\nWrite `solution.py` in the project root that reads from stdin and writes to stdout.\n\n---\n\n## Problem Overview\n\nRead `README.md` and `tools/README.md` for full problem details. A baseline `solution.py` already exists (it produces syntactically valid but low-quality output). Your job is to improve it.\n\n---\n\n## Evaluation\n\nYour solution is scored on **50 fixed test cases**. Final score = sum of individual case scores. **Higher is better.**\n\n---\n\n## Local Testing\n\nGenerate local random tests with `./tools/bin/gen <seed>`, using seeds in the range **0..10000** only.\n\n```bash\n# Generate a random test case (seed-based, deterministic)\n./tools/bin/gen 0 > input.txt\n\n# Run your solution\npython3 solution.py < input.txt > output.txt\n\n# Score output (Higher is better)\n./tools/bin/tester input.txt output.txt\n# Outputs to stderr: Score = <N>\n```\n\n---\n\n## Rules\n\n- Write your solution as `solution.py` in the project root directory\n- Do NOT modify files in `tools/`\n- Use `tools/bin/gen` and `tools/bin/tester` for local testing\n- For local scoring, use only `./tools/bin/tester`; do not use `tools/src/verifier.py` for scores\n- Your program should read from stdin and write to stdout\n- Run your solution to completion and verify with the tester before finishing\n"
  },
  "judge": {
    "image_tag": "d47dc1a7da74",
    "eval_cmd": "cd /home/workspace/treants_forest && python3 /tmp/eval_treants_forest.py",
    "eval_timeout": 870,
    "parser": "score_sum",
    "score_direction": "maximize",
    "selection": "score_first",
    "rescale": {
      "kind": "piecewise_max",
      "baseline": 5369.0,
      "rank30": 88189.0,
      "rank1": 330195.0,
      "super_anchor": 451198.0
    }
  }
}