The Agent Skills Directory

[SAFE]: The skill performs legitimate content transformation tasks. It reads local reference files and writes generated Discord announcements, thread prompts, and embed messages to local markdown files within the discord/ directory.
[PROMPT_INJECTION]: The skill demonstrates an indirect prompt injection surface by processing content 'atoms' that likely originate from external sources. However, the risk is minimal because the skill's capabilities are restricted to text generation and local file writing, with no access to sensitive system resources or network capabilities.
Ingestion points: Content atoms and thesis statements ingested via the atoms and main_argument parameters.
Boundary markers: Explicit delimiters for untrusted content are absent.
Capability inventory: Writing markdown files to the local file system (discord/announcement.md, discord/thread-prompt.md, discord/embed.md).
Sanitization: No input sanitization or validation logic is implemented.

repurpose-discord