The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references an external site, blog.xi-xu.me, as a primary conceptual source for its workflow. This domain is currently flagged as blacklisted by security scanners. While the domain name matches the author's handle, the blacklist status represents a potential risk if the site contains malicious content.
[PROMPT_INJECTION]: Automated scanners flagged potential concealment patterns in references/nginx-patterns.md. Manual analysis confirms these are false positives triggered by standard security terms like 'private' and 'loopback' used in the context of server hardening rather than an attempt to hide agent activities from the user.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it directs the agent to fetch and follow instructions from external live documentation sources.
Ingestion points: The skill relies on external documentation from well-known sources like ArchWiki, Nginx, and Ubuntu as specified in references/distro-routing.md.
Boundary markers: No delimiters or safety instructions are provided to help the agent distinguish between legitimate guidance and potential embedded malicious prompts.
Capability inventory: The agent generates system configuration and shell commands for users to execute.
Sanitization: There is no evidence of sanitization or validation for content retrieved from external documentation.
[NO_CODE]: The skill consists entirely of instructional text and configuration templates. It does not include any executable scripts, binaries, or automated tool calls that run within the agent's environment.

secure-linux-web-hosting