The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to fetch and load external documentation from https://www.pixazo.ai/models/qwen-audio.md at runtime to resolve parameter names and schemas.
Ingestion points: The reference section in SKILL.md directs the agent to load content from the vendor's documentation URL.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore or sanitize embedded commands within the fetched documentation.
Capability inventory: The skill is capable of executing network requests (via curl, Python requests, and Node.js fetch), performing file writes to the user's home directory (~/.pixazo/api-key), and modifying file permissions (chmod 600).
Sanitization: No sanitization or validation mechanisms are specified for the external content before it is processed by the agent.
[COMMAND_EXECUTION]: The skill utilizes shell commands for both API interactions and environment setup.
Evidence: Includes curl commands for making POST requests to the API gateway and chmod 600 to secure the locally stored API key file.
[EXTERNAL_DOWNLOADS]: Fetches supplementary model guidelines and technical specifications from the vendor's official website (www.pixazo.ai).

qwen-audio