Fraud Incident Extractor

Overview

A package designed to analyze user-submitted incident descriptions related to financial frauds, scams, or cybersecurity breaches. It processes the input text to extract structured details such as involved parties, amounts lost, scam types, and brief summaries.

Installation

pip install fraudincident_extractor

Usage

from fraudincident_extractor import fraudincident_extractor

user_input = "I lost $100 to a phishing scam. The scammer called me and asked for my bank details."

response = fraudincident_extractor(
    user_input=user_input,
    api_key="your_api_key",
    llm=ChatAnthropic()
)

print(response)

You can also use your own LLM instance from langchain by passing it like this:

from langchain_openai import ChatOpenAI
from fraudincident_extractor import fraudincident_extractor

llm = ChatOpenAI()
response = fraudincident_extractor(user_input=user_input, llm=llm)

or use anthropic:

from langchain_anthropic import ChatAnthropic
from fraudincident_extractor import fraudincident_extractor

llm = ChatAnthropic()
response = fraudincident_extractor(user_input=user_input, llm=llm)

or googl:

from langchain_google_genai import ChatGoogleGenerativeAI
from fraudincident_extractor import fraudincident_extractor

llm = ChatGoogleGenerativeAI()
response = fraudincident_extractor(user_input=user_input, llm=llm)

You can get a free API key for LLM7 by registering at https://token.llm7.io. If you want to use your own API key, you can pass it directly like this:

fraudincident_extractor(user_input=user_input, api_key="your_api_key")

You can also set the API key as an environment variable LLM7_API_KEY.

Contribution and Issues

If you encounter any issues or want to contribute to the package, please submit an issue to the GitHub repository: https://github.com/chigwell/fraud-incident-extractor

Author

Eugene Evstafev (chigwell) hi@euegne.plus