The Agent Skills Directory

[SAFE]: No security issues were detected. The skill primarily consists of markdown documentation and strategic guidelines for SaaS product management.
[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest a local context file (.agents/product-marketing-context.md) and process user-supplied feedback from exit surveys. While this constitutes a data ingestion surface, it is a documented and standard practice for this skill's persona as a SaaS consultant. Boundary markers are not explicitly defined, but the risk is assessed as safe given the lack of automated, high-privilege execution chains triggered by this data.
[COMMAND_EXECUTION]: The skill mentions various CLI tools (such as Stripe, PostHog, and Customer.io) as part of a tool registry for implementation. These are well-known, industry-standard services and their inclusion is consistent with the skill's stated purpose. There are no shell scripts or automated commands provided that would execute these tools without user oversight.

churn-prevention