churn-prevention
Pass
Audited by Gen Agent Trust Hub on Jul 15, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXPOSURE]: The skill instructs the agent to read local project files, specifically
.agents/product-marketing-context.mdor.claude/product-marketing-context.md, to gather business context before providing advice. This is an intended functional behavior for a context-aware assistant. - [INDIRECT_PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by ingesting data from the aforementioned local marketing context files. If these files are compromised or contain instructions from untrusted sources, they could influence the agent's behavior. However, the skill itself provides structured frameworks that mitigate accidental obedience to embedded instructions.
- [EXTERNAL_DOWNLOADS]: The skill references several well-known SaaS tools and billing providers (e.g., Stripe, Chargebee, PostHog, Customer.io) for implementation. These references are purely informational and point to well-known, trusted services.
Audit Metadata