DorkAgent

You can modify the number of results returned by the search engine by editing the tool source files within your environment's site-packages.

For Serper

Locate serper_dev_tool.py inside site-packages/crewai_tools/tools/serper_dev_tool/ and update the n_results attribute (range: 10 to 100).

For SerpApi

Locate serpapi_google_search_tool.py inside site-packages/crewai_tools/tools/serpapi_tool/. By default, it only forwards q and location. You must manually add a "num" key to the client.search({...}) dictionary within the _run() method to control the result count.

# Serper example (in serper_dev_tool.py)
n_results: int = 10 # min: 10, max: 100

# SerpApi example (in serpapi_google_search_tool.py)
# Add 'num' to the search dict in _run()
results = self.client.search({
    "q": kwargs.get("search_query"),
    "num": 50 
}).as_dict()

All functions, classes, and modules should include a single-line docstring that is concise and descriptive. Avoid multi-line docstrings unless necessary for complex logic, but the standard for DorkAgent is a single-line summary.

def adjust_depth(target_domains, depth):
    """Adjust domain search depth based on specified depth level"""
    pass

class DorkAgent:
    """Main agent class for Google Dorking operations"""
    pass

Use the following naming patterns to maintain consistency with the DorkAgent codebase:

• Functions: Use snake_case (e.g., send_notification).
• Variables: Use snake_case (e.g., target_domain).
• Constants: Use UPPERCASE_WITH_UNDERSCORES (e.g., MAX_RETRIES).
• Classes: Use PascalCase (e.g., DorkAgent).
• Files: Use lowercase with underscores (e.g., dorkagent_cli.py).

def send_notification(message):  # Good

target_domain = "example.com"    # Good

MAX_RETRIES = 3                  # Good

class DorkAgent:                # Good

# File: dorkagent_cli.py         # Good

DorkAgent is an LLM-powered agent for automated Google Dorking. To set it up, clone the repository and run the main script. The program handles auto-installation of required packages on its first run.

Prerequisites

• Python 3.11.x (specifically 3.11.9 is recommended).
• Note on Python versions: Using newer versions (like 3.13+) may cause installation failures for crewai dependencies (e.g., tiktoken) because prebuilt wheels are not yet available. If you encounter tiktoken build errors, downgrade to Python 3.11.x.

git clone https://github.com/yee-yore/DorkAgent.git
cd DorkAgent
python dorkagent.py

When running dorkagent.py for the first time, you will be prompted to select a search engine and an LLM provider. The tool will automatically save your keys to a .env file.

Required Search Engine Keys

You must provide one of the following depending on your runtime choice:

• SERPER_API_KEY: For Serper (obtain from serper.dev)
• SERPAPI_API_KEY: For SerpApi (obtain from serpapi.com)

Required LLM Keys

You must provide at least one LLM API key:

• OPENAI_API_KEY: For OpenAI models (GPT-4, GPT-5, etc.)
• ANTHROPIC_API_KEY: For Claude models
• GEMINI_API_KEY: For Gemini models

When adding imports to the DorkAgent project, organize them into three distinct groups in the following order:

1. Standard library imports.
2. Third-party library imports (e.g., crewai, langchain_openai, dotenv).
3. Local modules from the project itself.

Each group should be separated by a newline.

# 1. Standard library
import sys, re, os, pyfiglet
from datetime import datetime

# 2. Third-party libraries
from dotenv import load_dotenv
from crewai import Crew, LLM, Task, Agent
from langchain_openai import ChatOpenAI
from termcolor import colored
from prompt_toolkit import prompt
from prompt_toolkit.completion import PathCompleter
from crewai_tools import SerperDevTool, ScrapeWebsiteTool

# 3. Local modules
from local_module import function

To limit search results to a specific timeframe (e.g., the past week or month), use the tbs parameter in the search engine tools.

For Serper

In serper_dev_tool.py, modify the payload in _make_api_request using the tbs key:

• qdr:w: Past week
• qdr:m: Past month

For SerpApi

In serpapi_google_search_tool.py, add the "tbs" key to the client.search({...}) dictionary in the _run() method.

# Serper payload example
payload = json.dumps({"q": search_query, "num": self.n_results, "tbs": "qdr:m"})

# SerpApi search dict example
results = self.client.search({
    "q": kwargs.get("search_query"),
    "location": kwargs.get("location"),
    "tbs": "qdr:m",
}).as_dict()

Ensure proper spacing around operators and after colons. Avoid collapsing multiple statements onto a single line or omitting spaces in assignments.

# Good example
def verify_api_key(llm_type):
    """Verify API key validity"""
    required_keys = ["SERPER_API_KEY"]

    if llm_type == "openai":
        required_keys.append("OPENAI_API_KEY")

    return required_keys

# Bad example
def verify_api_key(llm_type):
    required_keys=["SERPER_API_KEY"]
    if llm_type=="openai": required_keys.append("OPENAI_API_KEY")
    return required_keys