The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch remote documentation and schema definitions from https://www.pixazo.ai/models/heygen.md to determine API parameters and polling shapes.
[COMMAND_EXECUTION]: The skill directs the agent to perform local file system operations, including creating a credential file at ~/.pixazo/api-key and modifying its permissions using chmod 600.
[COMMAND_EXECUTION]: The skill provides a suggested command npx skills add Pixazo-AI/skills which involves downloading and executing code from an external package registry.
[INDIRECT_PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it directs the agent to load and follow instructions from an external URL (https://www.pixazo.ai/models/heygen.md) to handle complex API interactions.
Ingestion points: SKILL.md (via explicit instructions to fetch heygen.md).
Boundary markers: None present; the agent is told to "load that URL when you need exact parameter names".
Capability inventory: Shell command execution (curl, chmod), network requests (requests.post, fetch), and external package execution (npx).
Sanitization: None present in the logic for processing the fetched documentation.

heygen