The Agent Skills Directory

[SAFE]: No security vulnerabilities were detected in the skill instructions or configuration files. The skill correctly implements role-based isolation to prevent context leakage during the adversarial marketing iteration loop. No risky capabilities such as network operations or arbitrary command execution are utilized.\n- [PROMPT_INJECTION]: The skill is designed to ingest and iterate on untrusted external data such as audience reviews and social media threads. This surface is managed through the use of isolated agent roles (Critic, Author, Synthesizer, Judge) and lacks any connection to sensitive tools or system-level capabilities, making the threat profile safe for intended marketing use cases.

marketing-autoreason