kelly-campaigns

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external subscriber data and campaign drafts from Busabase bases, which serves as an ingestion point for untrusted content. Ingestion points include the sends, segments, and suppression bases as handled in busabase-provider.js. Boundary markers are enforced through mandatory human review and explicit approval requirements defined in the SKILL.md instructions. Capability inventory includes writing records to the persistence layer and executing a handoff script. Sanitization is implemented using escapeHtml in the UI rendering logic across app.js and campaign-views.js.
  • [COMMAND_EXECUTION]: The skill includes a utility script, execute_decisions.mjs, which is intended to be executed locally via Node.js to finalize approved campaigns. The script re-validates quality gate verdicts and deliverability risks before marking records as ready for the external email service provider.
  • [EXTERNAL_DOWNLOADS]: The application communicates with busabase.com for data storage and references research documentation on mr-kelly.github.io. These destinations are recognized as official service endpoints and verified vendor-controlled repositories.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 10:36 AM
Security Audit — agent-trust-hub — kelly-campaigns