scripts/run_env.py

"""
scripts/run_env.py  — Run a single episode of the Self-Improving Agent Environment.

Usage
-----
python scripts/run_env.py [--task TASK_ID] [--seed SEED] [--verbose]

Examples
--------
python scripts/run_env.py
python scripts/run_env.py --task hard_coding_001 --seed 7 --verbose
"""

import argparse
import json
import logging
import sys
import os

# Make project root importable
sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))

from env.environment import SelfImprovingAgentEnv
from agent.baseline_agent import BaselineAgent


def main():
    parser = argparse.ArgumentParser(description="Run a single environment episode.")
    parser.add_argument("--task", type=str, default=None, help="Task ID to run (default: random)")
    parser.add_argument("--seed", type=int, default=42, help="Random seed (default: 42)")
    parser.add_argument("--verbose", action="store_true", help="Enable DEBUG logging")
    args = parser.parse_args()

    level = logging.DEBUG if args.verbose else logging.INFO
    logging.basicConfig(
        level=level,
        format="%(asctime)s [%(levelname)s] %(name)s: %(message)s",
    )

    env = SelfImprovingAgentEnv(seed=args.seed, task_id=args.task)
    agent = BaselineAgent(seed=args.seed)

    print("\n" + "=" * 60)
    print("  Self-Improving Tool-Optimizing Agent Environment")
    print("=" * 60)

    summary = agent.run_episode(env, task_id=args.task)

    print("\n📋 Episode Summary")
    print("-" * 40)
    for k, v in summary.items():
        print(f"  {k:20s}: {v}")

    print("\n🔄 Step-by-step history")
    print("-" * 40)
    for i, record in enumerate(agent.history, 1):
        action = record["action"]
        atype = action.get("action_type", "?")
        reward = record["reward"]
        info_keys = list(record["info"].keys())
        print(f"  Step {i:2d} | {atype:20s} | reward={reward:+.4f} | info keys={info_keys}")

    print("\n✅ Done. Log written to logs/env_log.jsonl")


if __name__ == "__main__":
    main()