The Agent Skills Directory

[SAFE]: The skill instructions provide standard marketing and conversion rate optimization (CRO) guidance without any prompt injection or bypass attempts.
[SAFE]: The Python script scripts/activation_funnel_analyzer.py performs local mathematical calculations on funnel data provided via JSON file or sample data. It does not perform network operations, execute shell commands, or use dangerous dynamic execution functions like eval() or exec().
[DATA_INGESTION]: The skill mentions reading .claude/product-marketing-context.md and the script reads a user-specified JSON file. These are standard ingestion patterns for context-aware agents and do not present a security risk as the processing logic is restricted to non-executable operations (prose generation and arithmetic).

cs-onboarding-cro