The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references and encourages the installation of the belt CLI tool and related skills from the inference-sh and belt-sh organizations. These downloads are associated with the platform's official infrastructure for running AI models.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-supplied prompts and image URLs which are then passed as arguments to the belt CLI tool. This creates a surface where malicious data could potentially influence the execution if not handled securely by the agent.
Ingestion points: Prompt text and image URLs provided in JSON payloads for the belt app run command.
Boundary markers: None defined in the skill instructions.
Capability inventory: Shell command execution limited to the belt binary through the Bash(belt *) tool restriction.
Sanitization: No explicit sanitization or validation steps are provided in the skill documentation.
[NO_CODE]: The skill consists entirely of Markdown instructions and documentation. It does not include any embedded scripts, configuration files, or binary executables that would run automatically.

flux-image