High severity · published Sep 15, 2026
CVE-2026-91943
Crawl4AI before 0.9.3 SSRF via PDFContentScrapingStrategy
7.7
CVSS
0.4%
EPSS
No
On KEV
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.
CVSS vector
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:N/A:N
Attack vector
Network
Complexity
Low
Privileges
Low
User interaction
None
Confidentiality
High
Integrity
None
Availability
None
Scope
Changed
Exploitation
Is it being used?
Exploit prediction (EPSS)0.4% · higher than 29% of CVEs
Exploited in the wildNo evidence yet
CISA KEVNot listed
Ransomware campaign useNot known
Affected products
Which versions
CR
crawl4ai · unclecode
before 0.9.3 · source: vendor
References
NVD
https://nvd.nist.gov/vuln/detail/CVE-2026-91943
CVE.org
https://www.cve.org/CVERecord?id=CVE-2026-91943
github.com
https://github.com/unclecode/crawl4ai/security/advisories/GHSA-q5rj-45vw-vp2g
www.vulncheck.com
https://www.vulncheck.com/advisories/crawl4ai-before-0.9.3-ssrf-via-pdfcontentscrapingstrategy
More from crawl4ai
10.0
CVE-2026-57572
Crawl4AI: Unauthenticated RCE via Chromium launch-argument injection in browser_config.extra_args
10.0
CVE-2026-53753
Crawl4AI: AST Sandbox Escape via gi_frame.f_back Chain - Pre-Auth RCE in Docker API
10.0
CVE-2026-26216
Crawl4AI < 0.8.0 Docker API Unauthenticated Remote Code Execution via Hooks Parameter
9.6
CVE-2026-57571
Crawl4AI arbitrary file write via download filename path traversal
8.6
CVE-2026-109238
Crawl4AI 0.7.0 before 0.9.4 SSRF via link_preview_config URL Seeder