The Agent Skills Directory

[SAFE]: The skill focuses on content strategy, SEO optimization, and editorial workflows. It provides detailed templates and frameworks for professional writing without introducing security risks.- [INDIRECT_PROMPT_INJECTION]: The skill includes modes for analyzing external data such as support tickets (Mode 11), competitor content (Mode 9), and user drafts (Mode 12). While this represents an inherent attack surface for indirect prompt injection, the instructions maintain a clear focus on structured analysis and creative production within the agent's context. No malicious exploitation or logic bypasses were identified.- [COMMAND_EXECUTION]: The README provides standard local installation instructions (mkdir, cp) to set up the skill in a local directory. These operations are transparent and involve no privileged escalation or dangerous shell commands.- [EXTERNAL_DOWNLOADS]: The skill mentions external tools for visual asset generation (e.g., code-as-images, video-generation) as optional hand-offs, but does not perform any unauthorized remote code downloads or execution at runtime.

content-writing