The Agent Skills Directory

[Prompt Injection] (SAFE): Analysis of the markdown files found no attempts to override agent instructions or bypass safety filters. The instructions are strictly focused on technical writing standards.\n- [Data Exposure & Exfiltration] (SAFE): No hardcoded credentials, sensitive file paths, or network exfiltration logic was found. The URLs provided in the documentation (e.g., Title Case Converter, TinyPNG) are informational and for manual user use.\n- [Unverifiable Dependencies & RCE] (SAFE): There are no package managers (npm, pip) or remote code execution patterns. The skill consists of static text content.\n- [Command Execution] (SAFE): While the string 'sudo apt install' appears in 'references/styleguide.md', it is explicitly presented as an example of how to write instructions for a reader, not as a command for the agent to execute.\n- [Indirect Prompt Injection] (LOW): The skill is designed to process untrusted user content (technical articles). However, its capabilities are limited to text analysis and feedback (display-only). There are no write operations or network calls that could be exploited via malicious content in the processed articles.

technical-writing-styleguide