kelly-app-skill-creator

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides a secure architectural framework for building Busabase-integrated agents without introducing malicious patterns, obfuscation, or unauthorized access vectors.
  • [DATA_EXFILTRATION]: The skill instructions enforce strict isolation of sensitive credentials. OAuth tokens are restricted to owner-only local files with 0600 permissions, and the system design ensures Vault secrets are never exposed to the browser or untrusted code environments.
  • [COMMAND_EXECUTION]: The skill implements security-conscious code patterns for generated applications, such as path-traversal guards for static file routing and centralized execution boundaries that separate UI logic from trusted backend operations.
  • [PROMPT_INJECTION]: To mitigate risks from processing external research data, the skill mandates human-in-the-loop review queues, deterministic decision records, and explicit sanitization and validation steps before data ingestion.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 08:00 AM
Security Audit — agent-trust-hub — kelly-app-skill-creator