The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references installation instructions hosted on GitHub.
Evidence: Links to https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md for CLI setup instructions.
[COMMAND_EXECUTION]: The skill instructs the agent to use a specific command-line interface (belt) for interacting with an external inference service.
Evidence: Multiple examples using belt app run to process images.
The allowed-tools metadata restricts execution to the belt command, adhering to the principle of least privilege.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data in the form of image URLs provided by the user or previous workflow steps.
Ingestion points: Image URLs passed to the --input argument in SKILL.md.
Capability inventory: Executes shell commands via the belt CLI.
Boundary markers: Not explicitly present in the command strings, though the command structure is well-defined.
Sanitization: Not explicitly detailed in the markdown instructions.

background-removal