Retire live model workflow E2E harnesses - #1578
Open
Open-Squilla wants to merge 1 commit into
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Scope
Retire the opt-in live model workflow E2E harnesses and their dedicated regression tests, coordinators, fault proxy, and browser driver. Remove the model-backed Gateway/search workflow steps and the live branches of the router and meta-skill validation scripts.
Preserve the three deterministic document Workbench flows in an offline test module. Their extracted helper accepts only a loopback provider and synthetic credentials. Update the desktop runner, Windows test inventory and balancing metadata, and contributor guidance to match.
Scope boundary: test infrastructure, CI metadata, and documentation.
Non-goals: product runtime changes; removal of standalone provider smoke tests, model evaluations, Telegram delivery checks, search-provider checks, or deterministic browser/desktop tests.
Branch
Base branch: main
Target exception: N/A
Issue
Linked issue: None
If None, reason: test infrastructure maintenance.
Release Note
Release note: NONE — no shipped behavior changes.
Tests
Ruff: all changed Python files passed.
Pytest: 304 passed, 1 skipped across the CI planner/workflow/shard, script, documentation, and public-release hygiene tests. Preserved offline Workbench flows: 2 passed, 1 skipped. Full collection: 26,034 tests collected successfully.
Build: not run; no runtime or frontend implementation changes. Full-suite execution and Linux/Windows runtime acceptance were not performed locally.
Regression tests: updated; preserve all three offline Workbench scenarios and their existing assertions. Make the CI planner regression use an isolated dependency fixture.
Notes: validated after rebasing onto current main and installing its two new locked dependencies. Existing skips require an xdist worker and hydrated Router Git LFS model assets respectively. Node syntax and diff whitespace checks passed. Windows balancing guardrails and retained historical duration values are unchanged.
The default test path remains offline, deterministic, credential-free, and safe for forks.
Maintainer Live Check
Maintainer live check: no
Surface: N/A
No real model API calls were made. Retiring these harnesses removes their manual full-flow evidence for real-provider routing, long tasks, recovery, skill generation, document editing, and search. The remaining provider smoke tests do not replace that coverage.
Safety
No product source, persisted configuration, database migration, or public client contract changes; no upgrade migration is required. The offline helper preserves portable child-process isolation, but only macOS loopback execution was verified locally. No secrets, runtime transcripts, private material, generated assets, or machine-specific paths are included.
Third-Party Origin
Third-party origin: none
Documentation Changes