openai-responses-adapter

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [NO_CODE]: The skill package is composed entirely of Markdown documentation, technical specifications, and a license file. It contains no executable scripts, binaries, or configuration files that would be executed by the agent environment, presenting no direct code execution risk.
  • [INDIRECT_PROMPT_INJECTION]: The skill describes a protocol for translating untrusted external request data (OpenAI Responses API) for upstream LLM gateways. This documentation outlines an attack surface where malicious input could influence a model, but provides robust mitigation strategies.
  • Ingestion points: External request data enters the system through the input and instructions fields of the POST /v1/responses endpoint, as detailed in SKILL.md and reference.md.
  • Boundary markers: The reference.md file explicitly states that the adapted chat body must be stripped of all Responses-specific keys before being passed to the gateway.
  • Capability inventory: The described system performs network operations (HTTPS) to forward requests to an inference gateway.
  • Sanitization: The specification defines a strict mapping and '400 vs strip' logic in SKILL.md to ensure that sensitive or unsupported fields are either rejected with an error or safely removed before reaching the LLM.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 04:17 PM
Security Audit — agent-trust-hub — openai-responses-adapter