The Agent Skills Directory

[SAFE]: The skill consists entirely of natural language instructions for reasoning. No scripts, binaries, or automated tool calls are included in the package.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-provided observations or explanations. While it lacks explicit delimiters to sanitize these inputs, the skill does not have access to any tools or file system operations, which restricts the potential impact of a prompt injection to the current conversation context.
[EXTERNAL_DOWNLOADS]: The documentation references the author's own GitHub repository for installation and updates. This is a standard vendor resource for distributing agent skills and does not involve untrusted third-party code execution within the skill's logic.

consider-alternatives