What problem does it solve? ToolUniverse tools fail integration tests due to JSON parsing errors, schema mismatches, invalid endpoints, wrong parameter types, and binary response handling issues, and diagnosing each failure manually is slow and error-prone. ## Core Features & Use Cases - Systematic Error Diagnosis: Classifies failures into nine error types (JSON parsing, schema validation, nullable fields, mixed types, invalid test examples, parameter errors, API key configuration, endpoint patterns, transient failures) with a fix location for each. - Targeted Fix Patterns: Provides concrete before/after code patterns for tool classes (*_tool.py) and JSON configs (*_tools.json), including binary response handling, nullable field schemas, and multi-operation data wrappers. - Test Validation Workflow: Covers integration tests, unit test updates in tests/unit/, and tool regeneration via python -m tooluniverse.generate_tools. - Use Case: A ChEMBL image endpoint fails with a JSON parsing error because the API returns SVG binary data; the skill guides adding Content-Type detection in the tool class and changing return_schema to {"type": "string"}, restoring a 100% pass rate. ## Quick Start Fix the failing ChEMBL_get_molecule_image tool by running its integration test, diagnosing the error, applying the appropriate fix, and verifying the tests pass.
