Nuclei CVE Generator

A Flask API that automatically generates Nuclei templates for CVEs using real-world PoC data from Exploit-DB and GitHub, powered by Claude AI.

---

How It Works

1. You send a CVE ID to the /generate endpoint
2. It searches Exploit-DB for matching exploits and fetches the raw exploit code
3. It searches GitHub for PoC repositories and fetches their README content
4. It builds a prompt from the real PoC data (HTTP requests, payloads, reproduction steps)
5. It sends the prompt to Claude and returns a ready-to-use Nuclei YAML template

---

Setup Options

• Option A — AWS Bedrock on EC2 (no API key needed, uses IAM roles)
• Option B — Anthropic API locally (simpler, just needs an API key)

---

Option A: Deploying on AWS EC2

1. Enable Bedrock Model Access

Before launching anything, make sure the Claude model is enabled in your AWS account:

1. Go to the AWS Bedrock console
2. Click Modify model access
3. Find Claude 3.5 Haiku under Anthropic and enable it
4. Wait for status to show Access granted (usually instant)

---

2. Create an IAM Role for EC2

Your EC2 instance needs permission to call Bedrock. The safest way is an IAM role — no hardcoded credentials needed.

1. Go to IAM → Roles → Create role
2. Select AWS service → EC2 → Next
3. Search for and attach the policy AmazonBedrockFullAccess (or create a custom policy scoped to just bedrock:InvokeModel)
4. Name the role something like nuclei-cve-bedrock-role and create it

---

3. Launch an EC2 Instance

1. Go to EC2 → Launch Instance
2. Choose Ubuntu Server 24.04 LTS (free tier eligible)
3. Instance type: t2.micro is fine for personal use, t3.small if you expect heavier load
4. Key pair: create or select an existing one — you'll need this to SSH in
5. Network settings:
   • Allow SSH (port 22) from your IP
   • Add a custom TCP rule for port 5000 from your IP (or 0.0.0.0/0 if you want it publicly accessible)
6. Under Advanced details → IAM instance profile: select the role you created in step 2
7. Launch the instance

---

4. Connect to the Instance

ssh -i /path/to/your-key.pem ubuntu@<your-ec2-public-ip>

If you get a permissions error on the key:

chmod 400 /path/to/your-key.pem

---

5. Set Up the Environment

# Update packages
sudo apt update && sudo apt upgrade -y

# Install Python and pip
sudo apt install python3 python3-pip python3-venv git -y

# Clone the repo
git clone https://github.com/CommandRunner/Nuclei_CVE_Generator
cd Nuclei_CVE_Generator

# Create and activate a virtual environment
python3 -m venv venv
source venv/bin/activate

# Install dependencies
pip install -r requirements.txt

---

6. Verify AWS Access

Since the EC2 instance has the IAM role attached, you don't need to store any credentials. Verify it's working:

sudo apt install awscli -y
aws sts get-caller-identity

You should see your account ID and the role ARN. If this works, Bedrock calls will work too.

---

7. Run the Server

python main.py

The API will be available at http://<your-ec2-public-ip>:5000.

---

8. Run as a Background Service (Optional)

sudo nano /etc/systemd/system/nuclei-cve.service

Paste the following (adjust paths if needed):

[Unit]
Description=Nuclei CVE Generator
After=network.target

[Service]
User=ubuntu
WorkingDirectory=/home/ubuntu/Nuclei_CVE_Generator
ExecStart=/home/ubuntu/Nuclei_CVE_Generator/venv/bin/python main.py
Restart=always

[Install]
WantedBy=multi-user.target

Then enable and start it:

sudo systemctl daemon-reload
sudo systemctl enable nuclei-cve
sudo systemctl start nuclei-cve
sudo systemctl status nuclei-cve

View logs:

journalctl -u nuclei-cve -f

---

Option B: Running Locally with the Anthropic API (No AWS Required)

If you don't want to set up AWS, you can swap Bedrock for the Anthropic API directly.

1. Get an Anthropic API Key

Sign up at console.anthropic.com and create an API key.

---

2. Install the Anthropic SDK

pip install anthropic

Or add it to requirements.txt:

anthropic>=0.25.0

---

3. Modify main.py

Replace the call_bedrock function with the following:

import anthropic

anthropic_client = anthropic.Anthropic(api_key=os.environ.get("ANTHROPIC_API_KEY"))

def call_bedrock(prompt):
    message = anthropic_client.messages.create(
        model="claude-3-5-haiku-20241022",
        max_tokens=4096,
        temperature=0.2,
        messages=[{"role": "user", "content": prompt}]
    )
    return message.content[0].text

You can remove the boto3 import and the bedrock client at the top of the file since they're no longer needed.

---

4. Set Your API Key and Run

git clone https://github.com/CommandRunner/Nuclei_CVE_Generator
cd Nuclei_CVE_Generator
pip install -r requirements.txt anthropic
export ANTHROPIC_API_KEY=your_api_key_here
python main.py

---

Usage

Generate a template:

curl -X POST http://localhost:5000/generate \
  -H "Content-Type: application/json" \
  -d '{"cve_id": "CVE-2021-44228"}'

Response:

{
  "template": "id: CVE-2021-44228\ninfo:\n  name: ...",
  "sources": {
    "exploitdb": [{"id": "...", "title": "...", "url": "..."}],
    "github": [{"repo": "...", "stars": 100, "url": "..."}]
  }
}

Increase GitHub API rate limit (optional):

export GITHUB_TOKEN=your_github_token

Without a token: 10 requests/min. With a token: 30 requests/min.

Health check:

curl http://localhost:5000/health

---

Security Notice

• Use only on authorized targets.
• Generated templates are AI-produced — always review before running against production systems.
• If exposing port 5000 publicly, consider putting it behind a reverse proxy (nginx) with authentication.