Skip to content

Security: LumenHelixLab/socialforge-ai

Security

SECURITY.md

Security Policy

Local Social Agent is an experimental local-first app. It is not production hardened.

Current safety posture

The app currently does not:

  • Store social media credentials
  • Publish to live social platforms
  • Auto-send posts
  • Execute browser automation
  • Expose a public multi-user service

Important safety rules

  • Raw creative cards must remain draft-only.
  • Raw cards must be promoted into safe reviewed cards before approval/scheduling.
  • Scheduling must not equal publishing.
  • Publishing must remain a separate future service with approval gates.
  • API credentials must never be exposed to AI agents directly.
  • Secrets must not be committed.

There aren't any published security advisories