← Back to search

CVE-2026-56712

8.5 HIGHpublic exploit available

Published 2026-11-09 · Updated 2026-11-09

AI risk analysis

Summary
Cog versions through 0.23.1 contain a server-side request forgery (SSRF) vulnerability in prediction inputs where attacker-controlled URLs in cog.File and cog.Path parameters are not validated against resolved host or IP before making HTTP requests. This allows authenticated attackers to force the Cog service to contact internal or sensitive endpoints, potentially exposing internal services or credentials.
Exploitability
Exploitation requires low skill and only valid user-level access to submit prediction inputs with malicious URLs; no authentication bypass is needed beyond standard login.
Blast radius
Successful exploitation could expose internal network resources, metadata services (e.g., AWS IMDS), or credential stores accessible from the Cog host, leading to data leakage or lateral movement within the cloud or on-premises infrastructure.
Detection
Defenders should monitor for outbound HTTP/HTTPS requests from the Cog service to non-public IP addresses (e.g., 127.0.0.0/8, 10.0.0.0/8, 172.16.0.0/12, 192.168.0.0/16) or to internal metadata endpoints (e.g., 169.254.169.254); such requests in logs from prediction workflows are strong indicators of SSRF attempts.
Prioritized remediation
Upgrade to Cog version 0.23.2 or later, which implements strict URL validation and blocks requests to internal IP ranges and localhost.
ssrfwebauth-bypasscloudinternal-exposure

Analysis generated locally by garnet-coder (no data left the box). AI-assisted — verify against primary sources before acting.

NVD description

Replicate Cog through 0.23.1 contains a server-side request forgery vulnerability that allows attackers to cause the Cog process to make server-side HTTP requests by submitting prediction inputs with attacker-controlled URLs that are not validated against the resolved destination host or IP before fetching. Attackers can supply malicious URLs or exploit HTTP redirects in cog.File and cog.Path predictor inputs to force requests to internal addresses, potentially exposing sensitive internal services or credentials.

CVSS vector

CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:L/A:N

Weaknesses

CWE-918

Public exploit & PoC references

All references

Source data: NVD (nvd.nist.gov), public domain. Exploit-DB.ai adds local AI analysis for defensive use only.

Related CVEs

Related by shared AI tags and CWE weakness class. Browse the full archive.