The Agent Skills Directory

[SAFE]: The skill serves as informational documentation for the Venice AI API, targeting the official vendor domain api.venice.ai. It follows security best practices by recommending the use of environment variables ($VENICE_API_KEY) for authentication headers.
[INDIRECT_PROMPT_INJECTION]: The skill describes methods for the agent to fetch external, potentially untrusted data such as public character descriptions and user reviews. This creates a surface for indirect prompt injection if that data is processed without caution.
Ingestion points: Data returned from GET /characters and GET /characters/{slug}/reviews (SKILL.md).
Boundary markers: The skill does not provide specific delimiters or instructions to the agent to disregard instructions found within the character data.
Capability inventory: The skill is purely documentation and does not provide or call any dangerous capabilities such as file-system writes, arbitrary shell execution, or network exfiltration tools.
Sanitization: No sanitization or filtering of the API response data is described.

venice-characters