The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill describes an API endpoint that ingests user-provided data, which constitutes an attack surface for indirect prompt injection if the processing agent does not apply sufficient guardrails.
Ingestion points: The input field in the POST /api/v1/responses request payload described in SKILL.md.
Boundary markers: The documentation does not explicitly define mandatory delimiters or boundary markers for content provided in the input field.
Capability inventory: The documented API supports text generation, reasoning processes, and tool/function execution.
Sanitization: No specific input sanitization or validation logic is detailed in the provided API reference.

venice-responses