list-growth-designer
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions include explicit security guardrails for handling untrusted data. In both the SKILL.md and skill-contract.md files, it specifically directs the agent to treat scraped text, reviews, and emails as evidence only and to ignore any embedded instructions within that content, effectively mitigating indirect prompt injection risks.\n- [CREDENTIALS_UNSAFE]: The skill references multiple external service connectors (e.g., HubSpot, Google Ads, Stripe, Mailchimp) but maintains a safe posture by explicitly prohibiting the storage of API keys within skill files or source control. No hardcoded secrets were found in the provided files.\n- [COMMAND_EXECUTION]: No dangerous shell commands, subprocess invocations, or privilege escalation attempts were detected. The skill's logic is focused on analysis and artifact generation within the agent's safe execution environment, with a preference for native MCP/connectors.
Audit Metadata