CVE-2026-95634: 
NixOS vulnerability analysis and mitigation

CVE-2026-95634 is a Server-Side Request Forgery (SSRF) vulnerability affecting Adobe's CAI Content Credentials suite, which could result in privilege escalation across scope boundaries. It affects the Content Credentials Rust SDK (c2pa) up to and including v0.90.15, C2PA Tool (c2patool) up to and including v0.27.15, and the Content Credentials JS SDK (c2pa-web) up to and including @contentauth/c2pa-web@0.13.4. The vulnerability was published on October 13, 2026, with patches made available the same day. It carries a CVSS v3.1 base score of 8.6 (High) (GitHub Advisory, Adobe Advisory).

The vulnerability is classified as CWE-918 (Server-Side Request Forgery), where the affected components fail to sufficiently validate or restrict URLs or similar requests before retrieving their contents, allowing an attacker to direct the server to make requests to unintended destinations. Exploitation requires no authentication, no user interaction, and has low attack complexity — an unauthenticated remote attacker can send crafted network requests to trigger the SSRF condition. The changed scope indicator in the CVSS vector reflects that the impact extends beyond the vulnerable component itself, enabling privilege escalation into adjacent systems or services (GitHub Advisory, Adobe Advisory).

Successful exploitation allows an unauthenticated remote attacker to access sensitive information from internal systems or services reachable by the vulnerable server, with high confidentiality impact. The changed scope means the attacker can pivot beyond the directly vulnerable component, potentially reaching internal metadata services (e.g., cloud instance metadata endpoints), internal APIs, or other network-accessible resources. Integrity and availability are not directly impacted, but the combination of information disclosure and privilege escalation could facilitate further attacks (GitHub Advisory, Adobe Advisory).

As of the time of disclosure, there is no public proof-of-concept exploit and no evidence of active in-the-wild exploitation (GitHub Advisory). The vulnerability is rated as automatable (no user interaction required), which increases the risk of mass exploitation if a PoC becomes available. The EPSS score is approximately 0.489%, placing it in the 41st percentile for exploitation likelihood within 30 days. The vulnerability is not currently listed in the CISA Known Exploited Vulnerabilities (KEV) catalog. Qualys has detection coverage for this vulnerability (detection ID 5019747) (GitHub Advisory).

1. Reconnaissance: Identify deployments of Adobe CAI Content Credentials components (c2patool, c2pa Rust SDK, or c2pa-web JS SDK) exposed to the network, particularly versions at or below c2patool-v0.27.15, c2pa-v0.90.15, or @contentauth/c2pa-web@0.13.4.
2. Craft SSRF payload: Construct a malicious request containing a URL pointing to an internal resource (e.g., cloud metadata endpoint http://169.254.169.254/latest/meta-data/ or internal service) that the vulnerable component will fetch without adequate validation.
3. Submit request: Send the crafted request to the vulnerable endpoint of the Content Credentials service without authentication, exploiting the lack of input validation on the URL parameter.
4. Retrieve sensitive data: The server fetches the attacker-specified URL and returns or processes the response, exposing internal credentials, tokens, or configuration data to the attacker.
5. Privilege escalation: Use retrieved credentials or tokens (e.g., cloud IAM tokens from metadata services) to escalate privileges within the environment or pivot to adjacent systems (GitHub Advisory, Adobe Advisory).

• Network: Outbound HTTP/HTTPS requests from the server running c2patool, c2pa, or c2pa-web to internal IP ranges (e.g., 169.254.169.254, 10.x.x.x, 172.16.x.x, 192.168.x.x) or unexpected external hosts; unusual DNS lookups for internal hostnames originating from the Content Credentials service process.
• Logs: Application or web server logs showing requests with URL parameters pointing to internal resources or cloud metadata endpoints; repeated or automated request patterns to Content Credentials API endpoints from a single source IP.
• Process: Unexpected network connections initiated by the c2patool or Node.js (c2pa-web) process to non-standard destinations; elevated process activity without corresponding user-initiated actions.

Adobe has released patched versions addressing this vulnerability: upgrade C2PA Tool to v0.28.1 or later, Content Credentials Rust SDK (c2pa) to v0.91.1 or later, and Content Credentials JS SDK (c2pa-web) to @contentauth/c2pa-web@0.15.3 or later (Adobe Advisory, GitHub Advisory). As a workaround prior to patching, restrict outbound network access from systems running these tools using firewall rules or network segmentation to block access to internal metadata services and sensitive internal endpoints. Organizations should also review and limit the network permissions of services running these SDKs to reduce the potential blast radius of SSRF exploitation.

The vulnerability received standard automated coverage across CVE tracking platforms and security news aggregators shortly after disclosure. AusCERT published a bulletin (ESB-2026.12622) and Fortress SRM included it in their October 2026 threat and security update. Social media activity was limited to automated CVE notification accounts on Bluesky. No notable independent researcher commentary or significant community debate has been observed beyond routine disclosure coverage (Adobe Advisory).