Skip to content

test(windows): stabilize shell suites and safe runner in WSL#92

Merged
T72 merged 1 commit intodevelopfrom
fix/tests-tm-agent-id-81
Mar 2, 2026
Merged

test(windows): stabilize shell suites and safe runner in WSL#92
T72 merged 1 commit intodevelopfrom
fix/tests-tm-agent-id-81

Conversation

@T72
Copy link
Copy Markdown
Owner

@T72 T72 commented Mar 2, 2026

Pull Request

Linked Issue

Closes #91

Related: #81, #82, #83, #84, #85, #86, #87, #88, #89, #90


Type

  • bug
  • refactor
  • feature
  • architecture
  • governance
  • infra
  • performance
  • security
  • test
  • documentation
  • verification

Summary

Stabilizes shell-based test suites and the safe test runner in WSL/Windows conditions by removing blocking sync calls and setting explicit non-interactive agent identity where orchestration enforcement requires it.


TDD Checklist (Required)

  • New behavior has tests (unit or contract).
  • Bug fixes include a regression test.
  • Refactors preserve behavior (tests protect it).
  • No skipped tests introduced.
  • CI is green.

If no tests were added, explain why:
The fixes are in test harness/scripts themselves. Validation is a full successful execution of the existing 12-suite safe runner after changes.


Architectural Integrity Checklist

  • No core-plugin coupling introduced.
  • No unintended public API contract changes.
  • Provider abstraction remains isolated.
  • Middleware ordering preserved (if modified).
  • Breaking change? (If yes, ADR required.)

Issue Hygiene Checklist

  • Linked issue exists and matches this PR scope.
  • If linked issue contains blocked-by, status labels are correct.

Acceptance Criteria Verification

  • Criterion 1: Full shell safe runner completes in WSL/Windows without hanging between suites. Satisfied by removing inter-suite sync from tests/run_all_tests_safe.sh.
  • Criterion 2: Suites that create tasks in non-interactive mode pass orchestration enforcement. Satisfied by setting TM_AGENT_ID in affected suites.
  • Criterion 3: Previously failing/hanging suites execute successfully with existing timeout budgets. Verified with targeted suite runs and full safe run.

Risk Notes (Optional)

  • Performance impact: none in product runtime; only test execution flow changed.
  • Cost impact: none.
  • Latency impact: test runtime more predictable due removal of blocking sync.
  • Migration required: none.

Additional Notes

Changed files:

  • tests/test_tm.sh
  • tests/test_edge_cases.sh
  • tests/additional_edge_tests.sh
  • tests/test_context_sharing.sh
  • tests/test_collaboration.sh
  • tests/test_agent_specialization.sh
  • tests/stress_test.sh
  • tests/test_durability.sh
  • tests/run_all_tests_safe.sh

End-to-end verification:

  • timeout 1200 bash tests/run_all_tests_safe.sh
  • Result: Total 12, Passed 12, Failed 0, Skipped 0

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

test-runner: remove blocking sync calls from remaining shell suites

1 participant