fix(e2e): restore WF12 hierarchical e2e test with OOM-safe implementation #11125

Closed
HAL9000 wants to merge 4 commits from test/restore-e2e-tests into master

4 Commits

Author SHA1 Message Date
freemo d6046fb0c1 fix(e2e): address review feedback from PR #11125 - resolve CI and review blockers
CI / helm (pull_request) Successful in 1m23s
CI / push-validation (pull_request) Failing after 1m54s
CI / build (pull_request) Successful in 2m42s
CI / lint (pull_request) Successful in 2m54s
CI / quality (pull_request) Successful in 3m13s
CI / typecheck (pull_request) Successful in 3m17s
CI / security (pull_request) Successful in 3m23s
CI / integration_tests (pull_request) Failing after 6m39s
CI / unit_tests (pull_request) Successful in 7m48s
CI / docker (pull_request) Successful in 1m49s
CI / coverage (pull_request) Successful in 10m46s
CI / status-check (pull_request) Failing after 3s
Applied fixes: duplicate CHANGELOG header removed, master.yml trailing newline added.

PR description updated to use 'Addresses' instead of 'Closes' for issue #10814 (Type/Testing).

(Note: tdd_quality_gate_steps.py split no longer needed — TDD quality gate was reverted on master.)
2026-05-13 10:48:08 +00:00
freemo 91730a7801 test: restore complete M2 acceptance e2e test
The M2 acceptance test was incomplete (truncated at line 48).
Restore the full test case that exercises the complete M2 milestone:
- Actor YAML compilation and registration
- Resource registration and project creation
- Action creation with custom actor configuration
- Full plan lifecycle: use, strategize, execute, diff, apply
- Verification of plan status and integrity

Exercises real LLM API integration using OpenAI GPT-4 model.

ISSUES CLOSED: #10812
2026-05-13 08:53:57 +00:00
freemo 1b0cec265e test: restore and enhance e2e test coverage
Restored deleted E2E test files features/cli_main_cov3.feature and
robot/plan_diff_artifacts.robot which were removed by commit 8ea00f51
(fix: restore CI quality tests to passing state). Updated CHANGELOG.md with
restoration entry and added contributor email.

ISSUES CLOSED: #8490
2026-05-13 08:53:57 +00:00
HAL9000 1003605cba fix(e2e): restore WF12 hierarchical e2e test with OOM-safe implementation
Restores the full WF12 large-scale hierarchical feature implementation E2E
test which was replaced by a TDD placeholder (tdd_expected_fail + Fail) in
commit 8ea00f51 due to plan execute being killed by SIGKILL (OOM) in CI.

OOM mitigations applied:
- Project repos use intentionally minimal single-line stub files (instead of
  multi-paragraph docstrings) to keep the LLM context window small during
  the strategize phase.
- Uses cautious automation profile (limited decomposition breadth) rather
  than supervised, reducing the scope of hierarchical decomposition.
- Removes tdd_expected_fail tag: the test now passes when the plan lifecycle
  completes successfully (the TDD placeholder is no longer needed).

Also updates stale command references:
- plan lifecycle-apply -> plan apply --yes (renamed in prior refactor)
- plan lifecycle-list -> plan list (renamed in prior refactor)

The tdd_issue and tdd_issue_4188 tags are retained as required by the TDD
workflow (bug fix PRs must keep tdd_issue_N tags without tdd_expected_fail).

ISSUES CLOSED: #10814
2026-05-13 08:47:39 +00:00