up

[Zochory]

This pull request removes several files related to Playwright and Codex auto-fix CI workflows, as well as a Python SSE performance test script. It also refactors the test suite to improve maintainability and test isolation, particularly in error handling and persistence tests.

CI and test infrastructure cleanup:

• Deleted .github/workflows/codex-autofix.yml, removing the Codex-based auto-fix workflow that attempted to automatically fix CI failures using OpenAI Codex.
• Removed Playwright configuration files playwright.config.ts and playwright.service.config.ts, eliminating browser-based end-to-end test setup and Azure-specific Playwright service configuration. [1] [2]

Test suite improvements:

• Refactored error handling tests in tests/test_error_handling.py to use shared constants for model names, reducing duplication and improving clarity. [1] [2] [3] [4] [5] [6] [7]
• Refactored persistence tests in tests/test_persistence.py to use a factory fixture for creating the persistence service, enabling customizable settings and better test isolation. All test cases now instantiate the service via the factory. [1] [2] [3] [4] [5] [6] [7]

Test script removal:

• Deleted test_sse_performance.py, removing a standalone script for testing SSE streaming performance and correlation tracking.## Description

[github-actions]

🤖 Hi @Zochory, I've received your request, and I'm working on it now! You can track my progress in the logs for more details.

[Copilot]

Pull Request Overview

This PR performs cleanup operations by removing unused scripts and configuration files, and improves test code quality through better fixture design and constant extraction.

• Removes obsolete tooling and workflow files (PyPI setup script, Playwright configs, Codex auto-fix workflow)
• Refactors test fixtures to use factory pattern for better flexibility
• Extracts magic strings into named constants for improved maintainability

Reviewed Changes

Copilot reviewed 8 out of 8 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
tools/scripts/setup-pypi-environment.sh	Deleted obsolete PyPI environment setup script
tests/validate_test_improvements.py	Updated comment to provide clearer guidance on handling import errors
tests/test_workflow_factory.py	Extracted validation logic into helper function, improved readability by replacing body with pass in skipped tests
tests/test_persistence.py	Converted fixture to factory pattern for customizable settings, improved summarization test assertions, replaced hardcoded model name with generic test value
tests/test_error_handling.py	Extracted magic strings into module-level constants, removed redundant assertion
playwright.service.config.ts	Deleted Azure Playwright service configuration file
playwright.config.ts	Deleted Playwright configuration file
.github/workflows/codex-autofix.yml	Removed Codex auto-fix workflow

[Copilot]

The constant MODEL_NAME is set to "agentic_fleet", but based on the codebase, the actual valid workflow/entity ID is "magentic_fleet" (note: "magentic" not "agentic"). This will cause all tests using MODEL_NAME to fail with entity_not_found errors since the entity doesn't exist. Change to MODEL_NAME = "magentic_fleet".

Suggested change

	MODEL_NAME = "agentic_fleet"
	MODEL_NAME = "magentic_fleet"

[Copilot]

The multi-condition check for summary messages combines three different possible field indicators (type, is_summary, content prefix), suggesting uncertainty about the actual schema. This makes the test fragile—if the implementation uses only one field consistently, the other checks are noise; if it varies, there's an underlying consistency issue. Either document which field(s) are canonical in a comment, or refactor to check only the fields actually used by ConversationPersistenceService.