The Agent Skills Directory

[SAFE]: The skill implements secure credential management by instructing the agent to store the Pixazo API key in a restricted file (~/.pixazo/api-key) with chmod 600 permissions, a recommended practice for local secrets.\n- [EXTERNAL_DOWNLOADS]: The skill fetches additional model reference documentation from the official vendor domain at https://www.pixazo.ai/models/idm-vton.md.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided garment descriptions and sends them to an external API, creating a potential surface for indirect prompt injection.\n
Ingestion points: User-supplied garment_des parameter in the API call request body in SKILL.md.\n
Boundary markers: Absent; the description is interpolated directly into the JSON payload.\n
Capability inventory: Network requests (HTTPS POST) to gateway.pixazo.ai using curl, Python requests, or Node.js fetch (SKILL.md), and file system write access for saving the API key (SKILL.md).\n
Sanitization: None; the skill does not specify filtering or escaping for the user-provided text.

idm-vton