Allow long OpenCode reviews

[seonghobae]

Summary

• restore the required OpenCode review job cap to GitHub's six-hour ceiling
• give the model-pool review step and opencode run a five-hour budget for large repositories
• keep the single nano model candidate and one attempt, so this changes patience rather than review breadth/cost

Verification

• py -3 -m pytest tests/test_opencode_agent_contract.py tests/test_opencode_workflow_shell_syntax.py tests/test_review_execution_contracts.py -q
• git diff --check -- .github/workflows/opencode-review.yml tests/test_opencode_agent_contract.py

[Copilot]

Pull request overview

This PR increases the allowed wall-clock time for the OpenCode PR review workflow to support large repositories by raising the job timeout to GitHub Actions’ 6-hour ceiling and extending the model-pool step/runtime budgets accordingly.

Changes:

• Increase the opencode-review-target job timeout from 75 minutes to 360 minutes.
• Increase the “Run OpenCode PR Review model pool” step timeout to 330 minutes and extend OpenCode runtime/retry budgets to 5 hours (18000 seconds).
• Update the OpenCode workflow contract test to assert the new timeout values.

Reviewed changes

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

File	Description
tests/test_opencode_agent_contract.py	Updates workflow contract assertions to match the new OpenCode job/step/runtime timeout budgets.
.github/workflows/opencode-review.yml	Raises job/step timeouts and increases opencode run / retry-budget environment values to allow longer review runs.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[github-actions]

Pull request overview

OpenCode reviewed the current-head evidence but found unresolved reviewer or review-agent threads before approval.

Findings

1. HIGH .github/workflows/opencode-review.yml:1 - Unresolved reviewer thread blocks automated approval

• Problem: OpenCode reached an APPROVE control result, but the approval step found unresolved, non-outdated human or review-agent thread evidence on the current pull request.
• Root cause: Reviewer and review-agent feedback can arrive after bounded model evidence is prepared, so the approval step must re-query GitHub immediately before publishing an approval.
• Fix: Address or resolve the listed reviewer thread(s), then re-run OpenCode on the current head.
• Regression test: Keep the approval gate querying reviewThreads(first: 100) after model output and before create_pull_review APPROVE, including bot review agents other than OpenCode itself.

Review thread evidence

Latest unresolved reviewer thread evidence

.github/workflows/opencode-review.yml line 2254

• Latest reviewer comment: @copilot-pull-request-reviewer at 2026-07-01T09:31:44Z

• Comment URL: #279 (comment)

• Comment excerpt: The Strix quick-gate regression script still hard-codes the previous OpenCode model-pool timeout settings (e.g., it asserts 'OPENCODE_RUN_TIMEOUT_SECONDS: "600"' and 'OPENCODE_TOTAL_RETRY_BUDGET_SECONDS: "3600"'). With these workflow values changed to 18000, 'scripts/ci/test_strix_quick_gate.sh' will fail when Strix runs its self-tests, blocking required checks.

• Result: REQUEST_CHANGES

• Reason: unresolved reviewer or review-agent thread(s) were present before approval.

• Head SHA: 5e2aee94ece7571f1b44a3dd9b40e3d751c58b05

• Workflow run: 28507529260

• Workflow attempt: 1

Changed-File Evidence Map

flowchart LR
PR["PR changed files"] --> Evidence["OpenCode bounded evidence"]
Evidence --> S1["Workflow: opencode-review.yml"]
S1 --> I1["GitHub Actions review job"]
I1 --> R1["Review risk: Workflow: opencode-review.yml"]
R1 --> V1["actionlint plus required checks"]
Evidence --> S2["Test: test_opencode_agent_contract.py"]
S2 --> I2["regression suite"]
I2 --> R2["Review risk: Test: test_opencode_agent_contract.py"]
R2 --> V2["targeted test run"]

Loading

[github-actions]

OpenCode Review Overview

• Head SHA: 4ede35e2b0bc0fab81faeddf2a8fb76a18a53906
• Workflow run: 28511481675
• Workflow attempt: 2
• Gate result: REQUEST_CHANGES (approval step)

Pull request overview

OpenCode reviewed the current-head evidence but found unresolved reviewer or review-agent threads before approval.

Findings

1. HIGH .github/workflows/opencode-review.yml:1 - Unresolved reviewer thread blocks automated approval

• Problem: OpenCode reached an APPROVE control result, but the approval step found unresolved, non-outdated human or review-agent thread evidence on the current pull request.
• Root cause: Reviewer and review-agent feedback can arrive after bounded model evidence is prepared, so the approval step must re-query GitHub immediately before publishing an approval.
• Fix: Address or resolve the listed reviewer thread(s), then re-run OpenCode on the current head.
• Regression test: Keep the approval gate querying reviewThreads(first: 100) after model output and before create_pull_review APPROVE, including bot review agents other than OpenCode itself.

Review thread evidence

Latest unresolved reviewer thread evidence

.github/workflows/opencode-review.yml line 2254

• Latest reviewer comment: @copilot-pull-request-reviewer at 2026-07-01T09:31:44Z

• Comment URL: Allow long OpenCode reviews #279 (comment)

• Comment excerpt: The Strix quick-gate regression script still hard-codes the previous OpenCode model-pool timeout settings (e.g., it asserts 'OPENCODE_RUN_TIMEOUT_SECONDS: "600"' and 'OPENCODE_TOTAL_RETRY_BUDGET_SECONDS: "3600"'). With these workflow values changed to 18000, 'scripts/ci/test_strix_quick_gate.sh' will fail when Strix runs its self-tests, blocking required checks.

• Result: REQUEST_CHANGES

• Reason: unresolved reviewer or review-agent thread(s) were present before approval.

• Head SHA: 4ede35e2b0bc0fab81faeddf2a8fb76a18a53906

• Workflow run: 28511481675

• Workflow attempt: 2

Changed-File Evidence Map

flowchart LR
PR["PR changed files"] --> Evidence["OpenCode bounded evidence"]
Evidence --> S1["Workflow: opencode-review.yml"]
S1 --> I1["GitHub Actions review job"]
I1 --> R1["Review risk: Workflow: opencode-review.yml"]
R1 --> V1["actionlint plus required checks"]
Evidence --> S2["CI script: test_strix_quick_gate.sh"]
S2 --> I2["review and security gate shell path"]
I2 --> R2["Review risk: CI script: test_strix_quick_gate.sh"]
R2 --> V2["bash -n plus Strix self-test"]
Evidence --> S3["Test: test_opencode_agent_contract.py"]
S3 --> I3["regression suite"]
I3 --> R3["Review risk: Test: test_opencode_agent_contract.py"]
R3 --> V3["targeted test run"]

Loading

[github-actions]

Pull request overview

OpenCode reviewed the current-head evidence but found unresolved reviewer or review-agent threads before approval.

Findings

1. HIGH .github/workflows/opencode-review.yml:1 - Unresolved reviewer thread blocks automated approval

• Problem: OpenCode reached an APPROVE control result, but the approval step found unresolved, non-outdated human or review-agent thread evidence on the current pull request.
• Root cause: Reviewer and review-agent feedback can arrive after bounded model evidence is prepared, so the approval step must re-query GitHub immediately before publishing an approval.
• Fix: Address or resolve the listed reviewer thread(s), then re-run OpenCode on the current head.
• Regression test: Keep the approval gate querying reviewThreads(first: 100) after model output and before create_pull_review APPROVE, including bot review agents other than OpenCode itself.

Review thread evidence

Latest unresolved reviewer thread evidence

.github/workflows/opencode-review.yml line 2254

• Latest reviewer comment: @copilot-pull-request-reviewer at 2026-07-01T09:31:44Z

• Comment URL: #279 (comment)

• Comment excerpt: The Strix quick-gate regression script still hard-codes the previous OpenCode model-pool timeout settings (e.g., it asserts 'OPENCODE_RUN_TIMEOUT_SECONDS: "600"' and 'OPENCODE_TOTAL_RETRY_BUDGET_SECONDS: "3600"'). With these workflow values changed to 18000, 'scripts/ci/test_strix_quick_gate.sh' will fail when Strix runs its self-tests, blocking required checks.

• Result: REQUEST_CHANGES

• Reason: unresolved reviewer or review-agent thread(s) were present before approval.

• Head SHA: 4ede35e2b0bc0fab81faeddf2a8fb76a18a53906

• Workflow run: 28511481675

• Workflow attempt: 2

Changed-File Evidence Map

flowchart LR
PR["PR changed files"] --> Evidence["OpenCode bounded evidence"]
Evidence --> S1["Workflow: opencode-review.yml"]
S1 --> I1["GitHub Actions review job"]
I1 --> R1["Review risk: Workflow: opencode-review.yml"]
R1 --> V1["actionlint plus required checks"]
Evidence --> S2["CI script: test_strix_quick_gate.sh"]
S2 --> I2["review and security gate shell path"]
I2 --> R2["Review risk: CI script: test_strix_quick_gate.sh"]
R2 --> V2["bash -n plus Strix self-test"]
Evidence --> S3["Test: test_opencode_agent_contract.py"]
S3 --> I3["regression suite"]
I3 --> R3["Review risk: Test: test_opencode_agent_contract.py"]
R3 --> V3["targeted test run"]

Loading

[seonghobae]

Pull request overview

OpenCode reviewed the current-head evidence but found unresolved reviewer or review-agent threads before approval.

Findings

1. HIGH .github/workflows/opencode-review.yml:1 - Unresolved reviewer thread blocks automated approval

• Problem: OpenCode reached an APPROVE control result, but the approval step found unresolved, non-outdated human or review-agent thread evidence on the current pull request.
• Root cause: Reviewer and review-agent feedback can arrive after bounded model evidence is prepared, so the approval step must re-query GitHub immediately before publishing an approval.
• Fix: Address or resolve the listed reviewer thread(s), then re-run OpenCode on the current head.
• Regression test: Keep the approval gate querying reviewThreads(first: 100) after model output and before create_pull_review APPROVE, including bot review agents other than OpenCode itself.

Review thread evidence

Latest unresolved reviewer thread evidence

.github/workflows/opencode-review.yml line 2254

• Latest reviewer comment: @copilot-pull-request-reviewer at 2026-07-01T09:31:44Z

• Comment URL: Allow long OpenCode reviews #279 (comment)

• Comment excerpt: The Strix quick-gate regression script still hard-codes the previous OpenCode model-pool timeout settings (e.g., it asserts 'OPENCODE_RUN_TIMEOUT_SECONDS: "600"' and 'OPENCODE_TOTAL_RETRY_BUDGET_SECONDS: "3600"'). With these workflow values changed to 18000, 'scripts/ci/test_strix_quick_gate.sh' will fail when Strix runs its self-tests, blocking required checks.

• Result: REQUEST_CHANGES

• Reason: unresolved reviewer or review-agent thread(s) were present before approval.

• Head SHA: 4ede35e2b0bc0fab81faeddf2a8fb76a18a53906

• Workflow run: 28511481675

• Workflow attempt: 2

Changed-File Evidence Map

flowchart LR
PR["PR changed files"] --> Evidence["OpenCode bounded evidence"]
Evidence --> S1["Workflow: opencode-review.yml"]
S1 --> I1["GitHub Actions review job"]
I1 --> R1["Review risk: Workflow: opencode-review.yml"]
R1 --> V1["actionlint plus required checks"]
Evidence --> S2["CI script: test_strix_quick_gate.sh"]
S2 --> I2["review and security gate shell path"]
I2 --> R2["Review risk: CI script: test_strix_quick_gate.sh"]
R2 --> V2["bash -n plus Strix self-test"]
Evidence --> S3["Test: test_opencode_agent_contract.py"]
S3 --> I3["regression suite"]
I3 --> R3["Review risk: Test: test_opencode_agent_contract.py"]
R3 --> V3["targeted test run"]

Loading

@copilot 고칩시다

[Copilot]

Pull request overview

OpenCode reviewed the current-head evidence but found unresolved reviewer or review-agent threads before approval....

해당 지적은 커밋 4ede35e에서 반영했습니다. scripts/ci/test_strix_quick_gate.sh의 검증값을 워크플로우와 동일하게 OPENCODE_RUN_TIMEOUT_SECONDS: "18000", OPENCODE_TOTAL_RETRY_BUDGET_SECONDS: "18000"로 맞췄습니다.

[seonghobae]

Closing as superseded. This PR only retunes the OpenCode model-pool knobs (OPENCODE_RUN_TIMEOUT_SECONDS 240→12600, OPENCODE_TOTAL_RETRY_BUDGET_SECONDS 360→12600, step timeout-minutes 75/20→360/240). main has since converged on a coherent, more evolved configuration (OPENCODE_RUN_TIMEOUT_SECONDS: "5400", OPENCODE_TOTAL_RETRY_BUDGET_SECONDS: "0", OPENCODE_MODEL_ATTEMPTS: "1", a 13-model candidate pool, model-pool step timeout-minutes: 350) via the newer opencode-review work (#350, #356, #360, #372). This PR's values conflict with that config and its intent ("allow long OpenCode reviews") is already delivered on main. No unique un-merged value remains.