An unauthenticated attacker can trigger a full remote code execution on any network‑accessible Paperclip instance running in authenticated mode with the default configuration. The vulnerability is achieved through an import authorization bypass that requires a chain of six API calls and no user interaction or credentials. Because the chain is fully automated against the default deployment, an adversary only needs the target’s address, resulting in a high‑impact breach of confidentiality, integrity, and availability. The weakness corresponds to a bypass of authentication and authorization controls (CWE‑287, CWE‑1188, CWE‑862).

The flaw affects the Paperclip AI platform, specifically paperclipai/server and paperclipai/paperclip services. All releases prior to version 2026.416.0 are vulnerable; the patch is included in the 2026.416.0 release.

The CVSS score of 10 classifies the scenario as critical, yet the EPSS score is less than 1%, indicating a low probability of seen exploitation at present. The exploit is network‑based and requires no credentials, making it an ideal target for automated attacks. Although the vulnerability is not listed in the CISA KEV catalog, its severity and the fully automated attack vector warrant immediate attention. A successful exploitation would grant the attacker arbitrary code execution on the host running Paperclip.