The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest untrusted external data (such as blog posts, podcast transcripts, and videos) and repurpose them into short-form content that can be automatically posted to social media.
Ingestion points: External content is processed in files like SKILL.md under conversion recipes for blogs, podcasts, and videos.
Boundary markers: There are no specific delimiters or instructions provided to the agent to treat external content as data rather than instructions.
Capability inventory: The skill utilizes the x/post-create app via the belt CLI to automate social media posts.
Sanitization: No explicit sanitization or validation of the processed text is mentioned before it is passed to the posting tool.
[COMMAND_EXECUTION]: The skill utilizes the Bash(belt *) tool to execute commands for media generation, transcription, and social media posting.
[EXTERNAL_DOWNLOADS]: The skill references and downloads external configurations and skills from GitHub and the inference.sh domain.
Evidence: The skill instructs the user to install the belt-sh/cli and inference-sh/skills packages and references a CLI installation markdown file hosted on the inference-sh GitHub repository.

content-repurposing