image_analyzer.py

import os
import base64
import requests
from smolagents import Tool

class ImageAnalysisTool(Tool):
    name = "image_analysis"
    description = "Analyze the content of an image and answer a specific question about it using Hugging Face Inference API."
    inputs = {
        "image_path": {
            "type": "string",
            "description": "Path to the image file (jpg, png, etc.)"
        },
        "question": {
            "type": "string",
            "description": "A question about the image content"
        }
    }
    output_type = "string"

    def __init__(self):
        super().__init__()
        api_token = os.getenv("HF_API_TOKEN")
        if not api_token:
            raise EnvironmentError("HF_API_TOKEN not found in environment variables.")
        self.api_url = "https://api-inference.huggingface.co/models/microsoft/git-base-captioning"
        self.headers = {
            "Authorization": f"Bearer {api_token}",
            "Content-Type": "application/json"
        }

    def forward(self, image_path: str, question: str) -> str