# CVE-2026-91943

## Summary

- **CVE ID:** CVE-2026-91943
- **Severity:** HIGH
- **CVSS Score:** 8.3 (CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:H/VI:N/VA:N/SC:H/SI:N/SA:N)
- **CWE:** CWE-918
- **Published:** Sep 15, 2026
- **Last Modified:** Sep 15, 2026

## Description

Crawl4AI before 0.9.3 contains a server-side request forgery vulnerability in PDFContentScrapingStrategy where _get_pdf_path() re-downloads targets with Python requests without egress validation. Authenticated attackers can supply URLs that redirect to internal addresses or use DNS rebinding to access internal services, exfiltrating responses through PDF text extraction in crawl results.

## Affected Products

- unclecode — crawl4ai (0)
- unclecode — crawl4ai (0.9.3)

## References

- [CNA](https://github.com/unclecode/crawl4ai/security/advisories/GHSA-q5rj-45vw-vp2g)
- [CNA](https://www.vulncheck.com/advisories/crawl4ai-before-0.9.3-ssrf-via-pdfcontentscrapingstrategy)

## Exploitation Prediction (EPSS)

- **EPSS Score:** 0.31%
- **EPSS Percentile:** 24.1

---
_Exported from OnDuty AI Vulnerability Intelligence on 2026-09-17._