The Agent Skills Directory

[SAFE]: The skill functions as an informational advisor for marketing optimization. It does not define any tools or execute any commands, and it contains no obfuscation or suspicious instruction patterns.\n- [SAFE]: The skill's behavior of reading local files like .agents/product-marketing.md for context is a standard, safe practice for tailoring agent output to specific projects. It does not attempt to access sensitive system files or credentials.\n- [INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface as it is designed to analyze user-provided marketing page content and URLs. However, it lacks any dangerous capabilities (such as network access, file writing, or command execution) that could be triggered by malicious instructions embedded in that data. The lack of functional capabilities makes this surface safe.\n- [REMOTE_CODE_EXECUTION]: No remote code execution patterns, package installations, or dynamic script generation were detected in the skill instructions or referenced documents.

cro