The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes the Bash tool to run a local script .claude/skills/ads/validate.py for CSV validation. This is an expected functional behavior for processing advertisement data.
[EXTERNAL_DOWNLOADS]: The skill fetches content from user-provided landing page URLs to perform 'Message Match' analysis. This is a primary feature of the skill's audit framework.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests untrusted data from external websites and CSV files.
Ingestion points: Processes data from user-supplied URLs and CSV export files (SKILL.md).
Boundary markers: None explicitly specified in the instructions to separate data from commands.
Capability inventory: The skill uses Read, Write, and Bash tools (SKILL.md).
Sanitization: No specific sanitization or validation of the external content is described in the prompt logic.

ads-review