The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references external resources including the installation of the belt-sh/cli via npx and provides links to setup instructions hosted on GitHub (inference-sh/skills). These are presented as legitimate dependencies for the skill's functionality.
[COMMAND_EXECUTION]: The skill uses the belt CLI within a Bash environment to run AI models and automate tasks like image generation and social media posting. All commands are transparent and align with the described purpose of the skill.
[PROMPT_INJECTION]: As the skill is designed to process external content (blogs, podcasts, transcripts) to generate derivative works, it has an inherent surface for indirect prompt injection.
Ingestion points: Content is ingested through command-line inputs or files provided to belt app run commands (e.g., transcripts for STT or blog text for thread generation).
Boundary markers: No specific delimiters or "ignore instructions" markers are used in the provided templates to isolate processed data from system instructions.
Capability inventory: The skill utilizes tools for file creation, network-based AI inference, and social media automation (x/post-create).
Sanitization: No explicit sanitization or input validation logic is present in the markdown instructions.

content-repurposing