The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the installation of the 'ai302' Python package from PyPI. This is a vendor-owned resource associated with the skill's primary functionality.
[COMMAND_EXECUTION]: The skill is centered around the execution of the 'ai302' CLI tool. It provides comprehensive instructions for the agent to construct and run commands for various media generation tasks.
[DATA_EXFILTRATION]: The CLI tool naturally communicates with the 302.AI API (302.ai) to process generation requests. This behavior is consistent with the stated purpose of the skill and involves sending prompts and media references to the vendor's infrastructure.
[PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes untrusted user prompts and external media (URLs or local files) which are then passed to the 'ai302' CLI. 1. Ingestion points: User-provided text prompts and file paths/URLs in commands like 'stt transcribe --file' or 'image create --image'. 2. Boundary markers: No explicit prompt boundaries or sanitization instructions are provided in the skill markdown. 3. Capability inventory: The agent has the capability to read local files, execute shell commands ('ai302'), and perform network operations via the CLI tool. 4. Sanitization: The instructions do not specify any sanitization or validation of the input content before processing.

ai302-cli