churn-prevention
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface area for processing untrusted external data via customer exit survey responses. \n
- Ingestion points: Free-text 'Other' reason fields and survey comments described in
SKILL.mdandreferences/cancel-flow-patterns.md. \n - Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions when the agent processes these customer responses. \n
- Capability inventory: The skill provides strategy and recommendations, referencing standard CLI tools like
stripe,customer-io, andposthogfor execution. \n - Sanitization: There are no instructions for sanitizing or escaping the content of survey responses before the agent evaluates them. \n- [COMMAND_EXECUTION]: The skill mentions the use of official CLI tools for well-known services such as Stripe, PostHog, and Customer.io. These tools are referenced as part of a legitimate integration registry for subscription management and analytics workflows.
Audit Metadata