oracle-family-scan

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and parses user-generated GitHub content from the public Soul-Brews-Studio/arra-oracle-v3 repo (e.g., SKILL.md Step 2 "Research" using gh issue view and the scan.sh / scripts/fleet-scan.ts flows that read issue/discussion titles/bodies), and then uses that content to craft and post personalized welcome messages, so untrusted third-party text can directly influence agent actions and outputs.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill runs gh API/GraphQL calls and gh issue view against the GitHub repo Soul-Brews-Studio/arra-oracle-v3 (https://github.com/Soul-Brews-Studio/arra-oracle-v3) at runtime to fetch issue bodies that are injected into the welcome-generation flow, so remote content directly controls the AI prompts.