paperclip

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The agent skill manages tasks, comments, and webhooks via the Paperclip API, where outsider-submitted issue comments and webhook payloads are ingested at runtime.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The referenced skills.sh URL is an external registry locator used for skill import, but the skills.sh service is an arbitrary third-party platform without official provenance.