scripts/run_agent.py

#!/usr/bin/env python3
"""
CLI: run the full CitationEdge pipeline on a PDF.

Usage:
    python scripts/run_agent.py path/to/paper.pdf
    python scripts/run_agent.py path/to/paper.pdf --job-id my_job_123
"""
import argparse
import asyncio
import json
import sys
from pathlib import Path

sys.path.insert(0, str(Path(__file__).parent.parent))

from dotenv import load_dotenv
load_dotenv()

from orchestrators.custom_orchestrator import CitationEdgeOrchestrator
from utils.helpers import generate_id
from utils.logger import get_logger

logger = get_logger("run_agent")


async def main():
    parser = argparse.ArgumentParser(description="Run CitationEdge analysis on a PDF")
    parser.add_argument("pdf_path", help="Path to the research paper PDF")
    parser.add_argument("--job-id", default=None, help="Custom job ID (auto-generated if omitted)")
    args = parser.parse_args()

    pdf_path = Path(args.pdf_path)
    if not pdf_path.exists():
        print(f"Error: PDF not found: {pdf_path}")
        sys.exit(1)

    job_id = args.job_id or generate_id("cli")
    print(f"\n=== Agentic CitationEdge ===")
    print(f"Job ID:   {job_id}")
    print(f"PDF:      {pdf_path}")
    print(f"Starting 7-agent pipeline...\n")

    orch = CitationEdgeOrchestrator()
    result = await orch.run(job_id=job_id, pdf_path=str(pdf_path))

    print("\n=== Pipeline Complete ===")
    print(json.dumps(result, indent=2, default=str))

    scores = result.get("scores", {})
    if scores:
        print(f"\n Overall Score: {scores.get('overall_score', 0):.1f}/10")
        print(f" Literary:     {scores.get('literary_score', 0):.1f}/10")
        print(f" Argument:     {scores.get('argument_score', 0):.1f}/10")
        print(f" Citations:    {scores.get('citation_completeness', 0):.1f}/10")

    print(f"\n Report: {result.get('report_path', 'N/A')}")


if __name__ == "__main__":
    asyncio.run(main())