response-compliance

The response-compliance skill is a specialized engineering utility designed for developers maintaining or extending the LobeHub Response API. It automates the execution of the official OpenResponses test suite, ensuring that your local or remote API implementations strictly adhere to required schemas and behavioral specifications. This tool is essential for CI/CD pipelines, debugging schema discrepancies, and validating core LLM interaction patterns before production deployment.

• Executes standardized test suites for the /api/v1/responses endpoint including basic text generation and streaming lifecycle verification.

• Validates complex response structures such as tool calling protocols, system prompt handling, multimodal image inputs, and multi-turn conversation history.

• Supports multiple authentication modes, including mock development users, API keys, and custom headers for flexible integration testing.

• Provides verbose output and JSON reporting capabilities, making it highly effective for both interactive debugging and automated quality assurance workflows.

• Automates the synchronization of the testing environment by cloning and updating the official openresponses test runner repository.

• Usage Note: Before running tests, ensure your local development server is running with ENABLE_MOCK_DEV_USER=true enabled.

• Input Requirements: Valid endpoint URL (e.g., http://localhost:3010) and appropriate authentication flags (e.g., --auth-header, --api-key).

• Failure Analysis: Use the -v flag to inspect full request/response payloads if tests fail due to Zod validation errors, missing output items, or auth redirect issues.

• Filter Tests: Use the --filter flag to target specific scenarios such as streaming-response, tool-calling, or system-prompt to isolate and fix specific bugs efficiently.