The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data fetched from external ad platforms, which presents a surface for indirect prompt injection if malicious content is present in campaign settings.
Ingestion points: Campaign metadata, ad group settings, and creative content are ingested from external APIs via scripts like google_ads_export_campaigns as described in SKILL.md.
Boundary markers: Absent. The instructions do not specify delimiters or instructions for the agent to ignore embedded commands within data retrieved from external sources.
Capability inventory: The skill can execute scripts to modify ad account settings, such as google_ads_set_campaign_geo_targets in SKILL.md.
Sanitization: Absent. No sanitization or safety filtering for the content of the retrieved campaign data is mentioned in the instructions.
[COMMAND_EXECUTION]: The skill utilizes a suite of validation scripts (e.g., google_ads_validate_campaign_structure, linkedin_ads_validate_campaign_structure) to interact with advertising platforms and perform technical audits.
[EXTERNAL_DOWNLOADS]: The documentation includes links to the author's official GitHub repository and service website for additional advertising automation tools. These are recognized as legitimate vendor resources.

campaign-preflight