## https://sploitus.com/exploit?id=D9A1D526-BF7C-5F9A-AC7B-2F541F5893F5
**PoC Finder โ Threat Intelligence โ PoC Search**
**Introduction**
This project aims to automatically convert threat intelligence (such as CVEs, product names, vulnerability descriptions, etc.) into queries for searching publicly available vulnerability databases (e.g., GitHub, Exploit-DB) to find PoCs (Proof-of-Concept). The retrieved results are then simply extracted and scored. This tool is intended only for defensive research, patch validation, and threat response. It must not be used for unauthorized attacks or malicious purposes. Quick start (Windows PowerShell).
1) Create and activate a virtual environment:
```powershell
python -m venv .venv; .\.venv\Scripts\Activate.ps1
```
2) Install dependencies:
```powershell
pip install -r requirements.txt
```
3) Set the GitHub Token (optional but recommended to increase the search limit):
```powershell
$env:GITHUB_TOKEN = "ghp_xxx"
```
4) Run the example script:
```powershell
python src/main.py --input sample_intel.json --output results.json
```
**File Structure**
- `src/main.py`: CLI entry point
- `src/intel_ingest.py`: Converts raw intelligence into searchable queries
- `src/search/github_search.py`: GitHub search adapter
- `src/search/exploitdb_search.py`: Simple interface for scraping Exploit-DB
- `src/poc_extractor.py`: Extracts PoC segments from the retrieved documents and scores them
- `prompts/prompts.md`: Various LLM prompts in English and Chinese, used to assist with searching, extraction, and evaluation
**Notes**
- This project is a prototype. It relies on public website structures, and scraping may fail due to changes in websites or anti-crawling mechanisms. For production environments, please use rate limits, proxies, and comply with the terms of service of the target sites. Please use the retrieved PoCs only within the scope of legal authorization.