social-sharing

Pass

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill involves fetching and parsing HTML and image metadata from web routes. This creates a data ingestion surface where the agent processes external content.
  • Ingestion points: Fetching resolved HTML and preview images from public or local routes as described in the 'Verify locally' and 'Verify after deployment' sections.
  • Boundary markers: None specified in the instructions.
  • Capability inventory: The skill utilizes file writing (to repair metadata) and command execution (running build and test scripts).
  • Sanitization: The instructions advise using URL APIs instead of string concatenation for encoding URLs, which is a positive security practice.
  • [COMMAND_EXECUTION]: The instructions direct the agent to execute standard project maintenance commands, including 'build', 'typecheck', and 'relevant tests' to ensure metadata changes do not break the application.
  • [SAFE]: The skill focuses on standard SEO and social sharing best practices and does not contain obfuscated code, credential harvesting, or unauthorized remote code execution patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 14, 2026, 06:57 AM
Security Audit — agent-trust-hub — social-sharing