Impact
Crawl4AI versions prior to 0.9.4 have a blind SSRF flaw in the RobotsParser.can_fetch routine. An unauthenticated attacker can supply a crafted robots.txt URL, enabling the crawler to reach arbitrary internal or cloud‑metadata addresses. This can expose sensitive information or enable further lateral movement without authentication. The weakness is classified as CWE‑918 for SSRF.
Affected Systems
The vulnerability affects the Crawl4AI project from the unclecode organization. Any installation running a version older than 0.9.4 is susceptible. Versions 0.9.4 and later contain a fix that disables the vulnerable behavior.
Risk and Exploitability
The CVSS score of 6.9 indicates moderate severity. No EPSS data is available, and the vulnerability is not listed in the CISA KEV catalog. Attackers likely need to direct the crawler to a maliciously controlled robots.txt file and craft redirects or DNS rebinding tricks to reach target internal services. Successful exploitation would allow the attacker to access internal services, gather sensitive information, and potentially compromise the confidentiality of those resources.
OpenCVE Enrichment