The Agent Skills Directory

REMOTE_CODE_EXECUTION (CRITICAL): The installation command curl -fsSL https://cli.inference.sh | sh downloads a script from an untrusted source and executes it directly in the system shell. This pattern is a major security vulnerability as the remote content is not cryptographically signed or verified before execution.
EXTERNAL_DOWNLOADS (HIGH): The skill uses npx skills add to fetch multiple packages (inference-sh/skills@...) from a source not included in the Trusted External Sources list. This introduces a supply chain risk where malicious code could be introduced via these external dependencies.
COMMAND_EXECUTION (MEDIUM): The skill utilizes a custom CLI tool (infsh) within Bash to perform operations like login, application execution, and social media posting. The skill requires full access to this tool, which can execute a wide variety of unverified remote AI applications.
PROMPT_INJECTION (LOW): The skill is vulnerable to indirect prompt injection through its core functionality of processing external data.
Ingestion points: The skill ingests untrusted long-form content such as blog posts and podcast transcripts.
Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are used when passing content to the conversion tools.
Capability inventory: The skill can execute commands, generate images/video, and post to social media platforms via the infsh CLI.
Sanitization: No evidence of sanitization or validation of the input content is provided before it is interpolated into tool commands.

content-repurposing