The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest external data from marketing-context.md and user-provided performance metrics. While this represents a potential attack surface, it is a standard functional requirement for marketing tools.
Ingestion points: marketing-context.md (referenced in SKILL.md) and user input for product, audience, and performance data.
Boundary markers: None explicitly defined in the instructions.
Capability inventory: The skill invokes scripts/ad_copy_validator.py for text analysis.
Sanitization: The skill performs regex-based compliance checks via the validator script but does not include explicit prompt-injection sanitization.
[COMMAND_EXECUTION]: The skill executes scripts/ad_copy_validator.py to ensure ad copy meets platform character limits and quality standards. The script is safe, utilizing only standard Python libraries (json, re, sys) without any network access or sensitive file operations.

ad-creative