The Agent Skills Directory

[SAFE]: The skill contains no executable code, shell scripts, or system commands. It functions as a set of logical instructions for the AI agent to follow within the conversation context.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted external content (the marketing copy to be optimized) and project-specific strategy files (.agents/marketing-strategy.md, .agents/marketing-learnings.md). While this represents a data ingestion surface, the risk is negligible because the skill does not possess high-privilege capabilities, such as network access or the ability to execute arbitrary commands, that could be abused by injected instructions.
[SAFE]: The skill includes a reference to an external GitHub repository for attribution purposes. This link targets a well-known service (GitHub) and is used neutrally for documentation rather than for remote code execution or data exfiltration.

autoresearch