totalrecon

totalrecon is a lightweight Python library for passive reconnaissance. It extracts subdomains, emails, and S3 buckets from text and PDF files, and uses a fine-tuned AI model to summarize sensitive infrastructure mentions.

Built for red teamers, bug bounty hunters, CTF players, and cyber analysts.

---

Warnings and Best Practices

Important:

This tool utilizes AI to assist with reconnaissance and summarization. While AI can provide valuable insights, always double-check the output. AI-generated results may not be 100% accurate and should be reviewed carefully before taking any action. The AI model used here is designed to assist, not replace, human judgment.

---

Features

• Extract intelligence from plaintext and PDF files
• Detect subdomains, emails, and AWS S3 buckets
• Summarize recon info with a fine-tuned flan-t5-small model
• Offline and lightweight — no OpenAI key required
• Trained on synthetic recon examples tailored for real-world use

---

Installation

pip install totalrecon

Or from source:

git clone https://github.com/josh1643/totalrecon.git
cd totalrecon
pip install .

---

Quick Start

Python Example

from totalrecon.extract import extract_from_text

text = '''
Found subdomain: api.dev.example.com
Email: admin@example.com
S3 bucket: s3://backup-prod-private
'''

results = extract_from_text(text)

print(results["domains"])          # ['api.dev.example.com']
print(results["emails"])           # ['admin@example.com']
print(results["s3_buckets"])       # ['s3://backup-prod-private']
print(results["recon_summaries"])  # ['Possible backup S3 bucket exposed via dev subdomain.']

---

About the Model

This project uses a fine-tuned FLAN-t5-small model hosted on the Hugging Face Hub:

🔗 https://huggingface.co/wassermanrjoshua/totalrecon-flan-t5

• Summarizes cyber recon and passive intel
• Runs entirely offline after first load
• No setup required — model is automatically downloaded on first use

This means:

• You don’t need to clone or manually download any model files
• Just pip install totalrecon and run it — the model loads when needed

---

Contributing

Contributions welcome!

1. Fork the repo
2. Create a feature branch
3. Open a pull request

---

License

MIT License — see LICENSE for full terms.

---

Author

Created by Joshua Wasserman for real-world recon workflows and open-source tooling.

• GitHub Repository: https://github.com/josh1643/totalrecon

---