social-presence

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill performs legitimate information gathering using official tools. It scans the user's own GitHub activity and public repositories associated with the author (harlan-zw) to suggest social media content.
  • [COMMAND_EXECUTION]: Uses the gh (GitHub) CLI to search for pull requests and releases. It specifically filters by the user's author tag (--author=@me) and fetches commit messages for specific repositories like nuxt-seo and unhead to provide context for tweet generation.
  • [DATA_EXFILTRATION]: No sensitive data access or unauthorized network transmissions were found. The skill logs its activity to ${CLAUDE_PLUGIN_DATA}/social-presence.log, which is the standard practice for plugin data persistence.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from GitHub PR titles and commit messages. While this presents a surface for indirect prompt injection (Category 8), the capabilities available to the skill are primarily read-only or limited to local logging, posing no significant risk of exploitation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 04:50 PM
Security Audit — agent-trust-hub — social-presence