The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read local project files, specifically .agents/product-marketing.md, .claude/product-marketing.md, or product-marketing-context.md, to gain context about the user's product. This creates a theoretical surface for indirect prompt injection if those files were to contain malicious instructions. However, the skill does not possess any capabilities (such as network access, file writing, or command execution) that would allow for an exploit or data exfiltration.
[SAFE]: The skill body and its references consist entirely of informational content and instructions for providing marketing advice. No obfuscation, hardcoded credentials, remote code execution patterns, or persistence mechanisms were found.

marketing-ideas