The Agent Skills Directory

[SAFE]: The skill serves as a configuration guide for an AI agent to interact with advertising APIs via Synter MCP. It does not contain any malicious code, hidden instructions, or obfuscated payloads.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external advertising performance data, which represents a potential injection surface if the ad data contains malicious instructions.
Ingestion points: Data retrieved through the pull_openai_ads_performance tool.
Boundary markers: The skill does not define specific text delimiters, but it explicitly requires human oversight.
Capability inventory: The agent can modify campaign states via update_campaign_budget and pause_campaign tools.
Sanitization: No explicit sanitization logic is provided in the markdown; however, the skill mandates a human-in-the-loop safety protocol ('never upload or enable without human review').
[EXTERNAL_DOWNLOADS]: The skill links to the author's official website (syntermedia.ai) and GitHub repository (github.com/Synter-Media-AI). These are recognized as legitimate vendor resources for the 'synter-media-ai' author.

openai-ads-operator