The Agent Skills Directory

[SAFE]: The skill is purely procedural and intended for text-based analysis and reporting. It does not use any tools, scripts, or system commands that could be exploited.
[INDIRECT_PROMPT_INJECTION]: The workflow involves processing external data (referred to as clusters and member posts) to generate section proposals. Although the skill does not define specific sanitization or boundary markers for this untrusted input, it possesses no functional capabilities (such as shell commands, network requests, or file-writing tools) that would allow malicious instructions within the data to impact the host system.

propose-section