The Agent Skills Directory

[Indirect Prompt Injection] (LOW): The skill processes user-provided audit data and context to generate social media posts. This creates a surface for indirect injection if the provided data contains malicious instructions. However, because the skill's capability is limited to generating text for the user to review and manually copy-paste, the threat is considered low risk.
[External Downloads] (SAFE): While the metadata mentions an external MCP server reference (@clawfu/mcp-skills), the skill itself does not execute any download commands or pull remote scripts.
[General Security] (SAFE): No evidence of prompt injection, data exfiltration, obfuscation, or persistence mechanisms was found. The skill operates within the expected boundaries of a content generation tool.

linkedin-post