The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill directs users to installation instructions on the author's GitHub organization (inference-sh) and utilizes tools from the author's official domain (inference.sh). These are identified as standard vendor resources.
[PROMPT_INJECTION]: The skill processes user-provided long-form content, creating a surface for indirect prompt injection. (1) Ingestion points: Content conversion recipes for blogs and podcast transcripts. (2) Boundary markers: Absent. (3) Capability inventory: Tool usage for posting to social media and generating assets via the 'belt' CLI. (4) Sanitization: Absent. This surface is characteristic of the skill's functionality and is considered safe.
[SAFE]: No evidence of credential exposure, malicious obfuscation, or unauthorized remote code execution was found.

content-repurposing