Fix OpenCode check lookup and review budget

[seonghobae]

Summary

• Extend the OpenCode model-pool job budget to 5 hours for large codebases.
• Keep the current single-model policy from main.
• Fix commit check-run pagination lookup by collecting pages first and applying jq locally, avoiding gh versions that reject --slurp with --jq.

Tests

• py -3 -m pytest tests\test_opencode_agent_contract.py tests\test_opencode_workflow_shell_syntax.py -q
• bash -n scripts/ci/test_strix_quick_gate.sh
• bash -n scripts/ci/run_opencode_review_model_pool.sh
• git diff --check

[Copilot]

Pull request overview

Adjusts the OpenCode review workflow’s model pool to reduce stalls by prioritizing faster OpenAI GitHub Models first and tightening timeout/retry limits, while keeping the “fail-closed on model exhaustion” behavior.

Changes:

• Reordered OPENCODE_MODEL_CANDIDATES to try OpenAI models before DeepSeek fallbacks.
• Reduced per-model run timeout (600s → 240s) and total retry budget (3600s → 1200s); reduced max backoff (600s → 120s).
• Updated contract tests to match the new timeout/budget values.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 2 comments.

File	Description
tests/test_opencode_agent_contract.py	Updates assertions to the new OpenCode run timeout and total retry budget values.
scripts/ci/test_strix_quick_gate.sh	Updates some workflow assertions for the new timeout/budget values (but still contains order-sensitive model-list assertions that now mismatch the workflow).
.github/workflows/opencode-review.yml	Reorders model candidates to OpenAI-first and reduces model-pool timeouts/retry budget/backoff limits.

---

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

[opencode-agent]

OpenCode Review Overview

• Head SHA: eb4dc12facccaeb22edf6324b079cb89d7bfb107
• Workflow run: 28535729994
• Workflow attempt: 1
• Gate result: REQUEST_CHANGES (approval step)

Pull request overview

OpenCode exhausted the configured model pool without a usable current-head review conclusion. This is not approval evidence, so the PR is blocked until a source-backed review can establish approval sufficiency or identify concrete fixes.

Findings

1. HIGH .github/workflows/opencode-review.yml:1 - OpenCode could not establish approval sufficiency

• Problem: every configured model path failed to produce a usable current-head control block.
• Root cause: model execution, timeout, export, normalization, or approval-gate validation did not complete after exponential retry across the configured model pool.
• Impact: approving from deterministic check state alone would miss PR-intent mismatches, missing files, edge-case bugs, robustness gaps, UX/DX regressions, security issues, and CodeGraph-backed base/head flow changes.
• Fix: rerun OpenCode after model availability recovers, or update the PR with the missing files, tests, docs, generated artifacts, and verification evidence needed for a source-backed review conclusion.
• Regression test: keep the approval gate posting REQUEST_CHANGES, not APPROVE or check-only failure, when no model produces a valid current-head review.

Summary

• Result: REQUEST_CHANGES
• Reason: coverage-evidence passed and peer GitHub Checks completed without failures, but no model produced a valid review control block.
• Deterministic evidence checked but not used for approval: current-head changed-file evidence (.github/workflows/opencode-review.yml, scripts/ci/test_strix_quick_gate.sh, tests/test_opencode_agent_contract.py); coverage-evidence result success; peer checks from statusCheckRollup excluding this OpenCode check.
• Model outcome: model_pool=exhausted; selected_model=none.
• Head SHA: eb4dc12facccaeb22edf6324b079cb89d7bfb107
• Workflow run: 28535729994
• Workflow attempt: 1

No PR approval was posted because model-output failure is not evidence that the PR has no blockers.

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 exhausted the configured model pool without a usable current-head review conclusion. This is not approval evidence, so the PR is blocked until a source-backed review can establish approval sufficiency or identify concrete fixes.

Findings

1. HIGH review evidence:1 - OpenCode could not establish approval sufficiency

• Problem: every configured model path failed to produce a usable current-head control block.
• Root cause: model execution, timeout, export, normalization, or approval-gate validation did not complete after exponential retry across the configured model pool.
• Impact: approving from deterministic check state alone would miss PR-intent mismatches, missing files, edge-case bugs, robustness gaps, UX/DX regressions, security issues, and CodeGraph-backed base/head flow changes.
• Fix: rerun OpenCode after model availability recovers, or update the PR with the missing files, tests, docs, generated artifacts, and verification evidence needed for a source-backed review conclusion.
• Regression test: keep the approval gate posting REQUEST_CHANGES, not APPROVE or check-only failure, when no model produces a valid current-head review.

Summary

• Result: REQUEST_CHANGES
• Reason: coverage-evidence passed and peer GitHub Checks completed without failures, but no model produced a valid review control block.
• Deterministic evidence checked but not used for approval: current-head changed-file evidence (.github/workflows/opencode-review.yml, scripts/ci/test_strix_quick_gate.sh, tests/test_opencode_agent_contract.py); coverage-evidence result success; peer checks from statusCheckRollup excluding this OpenCode check.
• Model outcome: model_pool=exhausted; selected_model=none.
• Head SHA: 62837fada2fefe39d23801ea6d62a88cfa76854b
• Workflow run: 28507945505
• Workflow attempt: 1

No PR approval was posted because model-output failure is not evidence that the PR has no blockers.

Inline comment note: OpenCode could not find an added RIGHT-side diff line for this PR, so the model-exhaustion blocker is attached to the PR review body instead of a file line.

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 exhausted the configured model pool without a usable current-head review conclusion. This is not approval evidence, so the PR is blocked until a source-backed review can establish approval sufficiency or identify concrete fixes.

Findings

1. HIGH review evidence:1 - OpenCode could not establish approval sufficiency

• Problem: every configured model path failed to produce a usable current-head control block.
• Root cause: model execution, timeout, export, normalization, or approval-gate validation did not complete after exponential retry across the configured model pool.
• Impact: approving from deterministic check state alone would miss PR-intent mismatches, missing files, edge-case bugs, robustness gaps, UX/DX regressions, security issues, and CodeGraph-backed base/head flow changes.
• Fix: rerun OpenCode after model availability recovers, or update the PR with the missing files, tests, docs, generated artifacts, and verification evidence needed for a source-backed review conclusion.
• Regression test: keep the approval gate posting REQUEST_CHANGES, not APPROVE or check-only failure, when no model produces a valid current-head review.

Summary

• Result: REQUEST_CHANGES
• Reason: coverage-evidence passed and peer GitHub Checks completed without failures, but no model produced a valid review control block.
• Deterministic evidence checked but not used for approval: current-head changed-file evidence (.github/workflows/opencode-review.yml, scripts/ci/test_strix_quick_gate.sh, tests/test_opencode_agent_contract.py); coverage-evidence result success; peer checks from statusCheckRollup excluding this OpenCode check.
• Model outcome: model_pool=exhausted; selected_model=none.
• Head SHA: 62837fada2fefe39d23801ea6d62a88cfa76854b
• Workflow run: 28507945505
• Workflow attempt: 1

No PR approval was posted because model-output failure is not evidence that the PR has no blockers.

Inline comment note: OpenCode could not find an added RIGHT-side diff line for this PR, so the model-exhaustion blocker is attached to the PR review body instead of a file line.

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 결정론적 리뷰 싫으니 exponential retry를 하든지 쫌

[Copilot]

Pull request overview

OpenCode exhausted the configured model pool without a usable current-head review conclusion. This is not approval evidence, so the PR is blocked until a source-backed re...

Fixed in commit that restores the full 13-model pool with OpenAI-first ordering and raises OPENCODE_BACKOFF_MAX_SECONDS from 30 → 600.

Root cause: the previous merge reduced OPENCODE_MODEL_CANDIDATES to a single entry (gpt-5-nano) with OPENCODE_MODEL_ATTEMPTS: "1", so there was exactly one attempt with no fallback — exhaustion was guaranteed on any transient failure.

Fix:

• OPENCODE_MODEL_CANDIDATES restored to the full pool: gpt-5 → gpt-5-chat → gpt-5-mini → gpt-5-nano → o3 → o3-mini → o4-mini → mistral-medium-2505 → llama-4-maverick → llama-4-scout → deepseek-r1-0528 → deepseek-r1 → deepseek-v3-0324 (OpenAI-first)
• OPENCODE_BACKOFF_MAX_SECONDS: "600" so per-model retry backoff can grow to 10 min instead of being capped at 30 s

With 1 attempt per model (keeping the "move through the catalog quickly" policy), 13 candidates cycle within the 5-hour budget rather than exhausting on the first failure.

[github-actions]

Pull request overview

OpenCode exhausted the configured model pool without a usable current-head review conclusion. This is not approval evidence, so the PR is blocked until a source-backed review can establish approval sufficiency or identify concrete fixes.

Findings

1. HIGH review evidence:1 - OpenCode could not establish approval sufficiency

• Problem: every configured model path failed to produce a usable current-head control block.
• Root cause: model execution, timeout, export, normalization, or approval-gate validation did not complete after exponential retry across the configured model pool.
• Impact: approving from deterministic check state alone would miss PR-intent mismatches, missing files, edge-case bugs, robustness gaps, UX/DX regressions, security issues, and CodeGraph-backed base/head flow changes.
• Fix: rerun OpenCode after model availability recovers, or update the PR with the missing files, tests, docs, generated artifacts, and verification evidence needed for a source-backed review conclusion.
• Regression test: keep the approval gate posting REQUEST_CHANGES, not APPROVE or check-only failure, when no model produces a valid current-head review.

Summary

• Result: REQUEST_CHANGES
• Reason: coverage-evidence passed and peer GitHub Checks completed without failures, but no model produced a valid review control block.
• Deterministic evidence checked but not used for approval: current-head changed-file evidence (.github/workflows/opencode-review.yml, scripts/ci/test_strix_quick_gate.sh, tests/test_opencode_agent_contract.py); coverage-evidence result success; peer checks from statusCheckRollup excluding this OpenCode check.
• Model outcome: model_pool=exhausted; selected_model=none.
• Head SHA: eb4dc12facccaeb22edf6324b079cb89d7bfb107
• Workflow run: 28528205754
• Workflow attempt: 1

No PR approval was posted because model-output failure is not evidence that the PR has no blockers.

Inline comment note: OpenCode could not find an added RIGHT-side diff line for this PR, so the model-exhaustion blocker is attached to the PR review body instead of a file line.

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 exhausted the configured model pool without a usable current-head review conclusion. This is not approval evidence, so the PR is blocked until a source-backed review can establish approval sufficiency or identify concrete fixes.

Findings

1. HIGH .github/workflows/opencode-review.yml:1 - OpenCode could not establish approval sufficiency

• Problem: every configured model path failed to produce a usable current-head control block.
• Root cause: model execution, timeout, export, normalization, or approval-gate validation did not complete after exponential retry across the configured model pool.
• Impact: approving from deterministic check state alone would miss PR-intent mismatches, missing files, edge-case bugs, robustness gaps, UX/DX regressions, security issues, and CodeGraph-backed base/head flow changes.
• Fix: rerun OpenCode after model availability recovers, or update the PR with the missing files, tests, docs, generated artifacts, and verification evidence needed for a source-backed review conclusion.
• Regression test: keep the approval gate posting REQUEST_CHANGES, not APPROVE or check-only failure, when no model produces a valid current-head review.

Summary

• Result: REQUEST_CHANGES
• Reason: coverage-evidence passed and peer GitHub Checks completed without failures, but no model produced a valid review control block.
• Deterministic evidence checked but not used for approval: current-head changed-file evidence (.github/workflows/opencode-review.yml, scripts/ci/test_strix_quick_gate.sh, tests/test_opencode_agent_contract.py); coverage-evidence result success; peer checks from statusCheckRollup excluding this OpenCode check.
• Model outcome: model_pool=exhausted; selected_model=none.
• Head SHA: eb4dc12facccaeb22edf6324b079cb89d7bfb107
• Workflow run: 28535729994
• Workflow attempt: 1

No PR approval was posted because model-output failure is not evidence that the PR has no blockers.

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]

HIGH OpenCode could not establish approval sufficiency

• Problem: the model pool exhausted without a valid current-head review control block, so this changed line cannot be approved from deterministic check state alone.
• Impact: PR-intent mismatches, missing files, robustness bugs, UX/DX regressions, and CodeGraph-backed flow changes could be missed.
• Fix: rerun OpenCode after model availability recovers, or add the missing source/test/docs/generated verification evidence needed for a source-backed approval.
• Verification: rerun the OpenCode Review workflow and confirm it emits APPROVE or source-backed REQUEST_CHANGES for this head SHA.

[seonghobae]

Closing as superseded. This PR only retunes model-pool knobs (RUN_TIMEOUT 240→17400, TOTAL_RETRY_BUDGET 360→18000, BACKOFF_MAX 30→600, step timeout-minutes 20→300, expanded candidate list) plus a --slurp→jq -s rewrite of the commit check-runs lookup. main converged on RUN_TIMEOUT=5400/RETRY_BUDGET=0/ATTEMPTS=1/13-model pool/step 350m via #350/#356/#360/#372, and main's contract tests now REQUIRE the --slurp | jq -r "$jq_filter" pattern this PR removes. Competing values conflict with main; intent already delivered. No unique un-merged value remains.