The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external content to automate social media posts and media creation, creating a risk if that content contains malicious instructions.\n
Ingestion points: The skill is designed to ingest blog posts, podcast transcripts, and video content as described in SKILL.md.\n
Boundary markers: There are no instructions or delimiters provided to isolate the source content or prevent the agent from following embedded prompts within that content.\n
Capability inventory: The skill has the ability to post content to social media platforms through the belt app run x/post-create command and generate various media files (SKILL.md).\n
Sanitization: No content sanitization, filtering, or validation steps are defined in the instructions to handle potentially adversarial input.\n- [EXTERNAL_DOWNLOADS]: The skill references external resources for installation and setup.\n
Evidence: The skill provides a link to an external markdown file at https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md for CLI installation instructions.\n- [COMMAND_EXECUTION]: The skill relies on a specific command-line interface tool for its primary operations.\n
Evidence: SKILL.md contains numerous Bash code blocks executing the belt command to interface with remote applications.

content-repurposing