feat(opencode): add empirical review-quality gate

.github/workflows/opencode-review-quality-ci.yml

@@ -0,0 +1,89 @@
+name: OpenCode Review Quality CI
+
+on:
+  pull_request:
+    branches: [main]
+    paths:
+      - ".github/workflows/opencode-review-quality-ci.yml"
+      - "benchmarks/opencode_review/**"
+      - "scripts/ci/opencode_review_quality_score.py"
+      - "tests/test_opencode_review_quality_score.py"
+      - "tests/test_opencode_review_quality_evidence_contract.py"
+      - "docs/doctoring/opencode-review-quality-evaluation.md"
+      - "docs/superpowers/specs/2026-08-08-opencode-review-quality.md"
+      - "docs/superpowers/plans/2026-08-08-opencode-review-quality.md"
+      - "CHANGELOG.md"
+
+permissions:
+  contents: read
+
+concurrency:
+  group: opencode-review-quality-${{ github.event.pull_request.number || github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  empirical-review-quality:
+    name: empirical-review-quality
+    if: github.event_name != 'pull_request' || github.event.action != 'closed'
+    runs-on: ubuntu-24.04
+    timeout-minutes: 10
+    steps:
+      - name: Checkout exact source revision
+        uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
+        with:
+          ref: ${{ github.event.pull_request.head.sha || github.sha }}
+          persist-credentials: false
+
+      - name: Set up Python
+        uses: actions/setup-python@5fda3b95a4ea91299a34e894583c3862153e4b97 # v7.0.0
+        with:
+          python-version: "3.14"
+
+      - name: Install exact hash-verified test runner dependencies
+        env:
+          PIP_DISABLE_PIP_VERSION_CHECK: "1"
+          PIP_NO_INPUT: "1"
+        shell: bash --noprofile --norc -e -o pipefail {0}
+        run: |
+          cat >"${RUNNER_TEMP}/opencode-review-quality-requirements.txt" <<'EOF'
+          coverage==7.15.2 --hash=sha256:b9a6367e4aff723e8ee8190836836124284e8fcd4265e307c844010cfa074f3f
+          iniconfig==2.1.0 --hash=sha256:9deba5723312380e77435581c6bf4935c94cbfab9b1ed33ef8d238ea168eb760
+          packaging==26.2 --hash=sha256:5fc45236b9446107ff2415ce77c807cee2862cb6fac22b8a73826d0693b0980e
+          pluggy==1.6.0 --hash=sha256:e920276dd6813095e9377c0bc5566d94c932c33b27a3e3945d8389c374dd4746
+          pygments==2.20.0 --hash=sha256:81a9e26dd42fd28a23a2d169d86d7ac03b46e2f8b59ed4698fb4785f946d0176
+          pytest==9.1.1 --hash=sha256:37a86b45efb9a47a61a36449063e8e18d0cab3161329fc099eb21783169c4f0c
+          EOF
+          python -m pip install \
+            --only-binary=:all: \
+            --require-hashes \
+            -r "${RUNNER_TEMP}/opencode-review-quality-requirements.txt"
+
+      - name: Verify empirical review quality contracts
+        shell: bash --noprofile --norc -e -o pipefail {0}
+        run: |
+          test "$(git rev-parse HEAD)" = "${{ github.event.pull_request.head.sha || github.sha }}"
+          python -m coverage run \
+            --branch \
+            --source=scripts/ci \
+            -m pytest \
+            tests/test_opencode_review_quality_score.py \
+            tests/test_opencode_review_quality_evidence_contract.py \
+            -q
+          python -m coverage report \
+            --include='scripts/ci/opencode_review_quality_score.py' \
+            --fail-under=100 \
+            --show-missing
+          python scripts/ci/opencode_review_quality_score.py \
+            --input benchmarks/opencode_review/pilot_baseline_v1.json \
+            --json-output "${RUNNER_TEMP}/opencode-review-quality.json" \
+            --markdown-output "${RUNNER_TEMP}/opencode-review-quality.md"
+          python -m json.tool "${RUNNER_TEMP}/opencode-review-quality.json" >/dev/null
+          python -m json.tool benchmarks/opencode_review/pilot_baseline_v1.json >/dev/null
+          grep -Fq 'Parity gate: **INSUFFICIENT_EVIDENCE**' \
+            "${RUNNER_TEMP}/opencode-review-quality.md"
+          cat "${RUNNER_TEMP}/opencode-review-quality.md" >>"$GITHUB_STEP_SUMMARY"
+          python -m compileall -q \
+            scripts/ci/opencode_review_quality_score.py \
+            tests/test_opencode_review_quality_score.py \
+            tests/test_opencode_review_quality_evidence_contract.py
+          git diff --exit-code

CHANGELOG.md

@@ -8,6 +8,7 @@ Semantic Versioning where the repository publishes a release.
 
 ### Added
 
+- Added an empirical OpenCode review-quality benchmark, fail-closed scorer, exact-head quality workflow, and APA 7th doctoring that keep lifecycle-yield evidence separate from head-matched expert-gold precision and recall, require Wilson-bound non-inferiority before any CodeRabbit-parity claim, and preserve 100% production statement/branch/docstring evidence.
 - Added a trusted pull-request comment router for `@cwl-noema-review` and review-only `@opencode-agent` dispatches, with an organization sweep, exact-head receipts, repository allowlisting, fixed runners, immutable checkout pins, and a permanent 100% statement/branch/docstring quality gate.
 - Added exact-base `uv.lock` materialization that reconstructs standalone nested projects with a checksum-pinned official `uv` exporter, isolated frozen/offline execution, strict exact-pin and SHA-256 output validation, and complete Python 3.10/3.14 quality evidence.
 

benchmarks/opencode_review/pilot_baseline_v1.json

@@ -0,0 +1,186 @@
+{
+  "schema_version": "1.0",
+  "benchmark_id": "cwl-opencode-lifecycle-pilot-2026-08-08",
+  "evaluation_mode": "historical_lifecycle",
+  "limitations": [
+    "The three pull requests were deliberately selected because both OpenCode and CodeRabbit evidence was observable; this is not a random organization-wide sample.",
+    "Reviewer outputs were produced on different lifecycle heads, so actionable yield is an operational contrast rather than a head-matched precision or recall comparison.",
+    "Actionable means that the reviewer emitted a concrete source-oriented finding; this pilot does not label each finding as an expert-verified true positive.",
+    "CodeRabbit rate-limited evidence is counted only when a GitHub record was directly observed, so availability may be overestimated.",
+    "No parity claim is permitted until a stratified head-matched expert-gold benchmark satisfies the configured sample-size floor."
+  ],
+  "parity_policy": {
+    "candidate_reviewer": "opencode",
+    "reference_reviewer": "coderabbit",
+    "minimum_head_matched_cases": 50,
+    "minimum_gold_findings": 50,
+    "non_inferiority_margin": 0.05,
+    "required_critical_high_recall": 1.0
+  },
+  "cases": [
+    {
+      "case_id": "disksage-pr-140",
+      "repository": "ContextualWisdomLab/disksage",
+      "pull_request_number": 140,
+      "head_match": false,
+      "diff_size_bucket": "small",
+      "primary_language": "rust",
+      "gold_findings": [],
+      "reviewers": {
+        "opencode": {
+          "triggered_attempts": 4,
+          "completed_attempts": 4,
+          "rate_limited_attempts": 0,
+          "infrastructure_only_reviews": 4,
+          "duplicate_reviews": 3,
+          "findings": []
+        },
+        "coderabbit": {
+          "triggered_attempts": 2,
+          "completed_attempts": 1,
+          "rate_limited_attempts": 1,
+          "infrastructure_only_reviews": 0,
+          "duplicate_reviews": 0,
+          "findings": [
+            {
+              "finding_id": "coderabbit-disk-140-publish-boundary",
+              "gold_finding_id": null,
+              "severity": "medium",
+              "actionable": true,
+              "source_backed": true,
+              "line_anchored": true,
+              "has_fix_direction": true,
+              "has_regression_test_direction": true
+            }
+          ]
+        }
+      }
+    },
+    {
+      "case_id": "egressweave-pr-62",
+      "repository": "ContextualWisdomLab/EgressWeave",
+      "pull_request_number": 62,
+      "head_match": false,
+      "diff_size_bucket": "medium",
+      "primary_language": "python",
+      "gold_findings": [],
+      "reviewers": {
+        "opencode": {
+          "triggered_attempts": 2,
+          "completed_attempts": 2,
+          "rate_limited_attempts": 0,
+          "infrastructure_only_reviews": 2,
+          "duplicate_reviews": 1,
+          "findings": []
+        },
+        "coderabbit": {
+          "triggered_attempts": 1,
+          "completed_attempts": 1,
+          "rate_limited_attempts": 0,
+          "infrastructure_only_reviews": 0,
+          "duplicate_reviews": 0,
+          "findings": [
+            {
+              "finding_id": "coderabbit-egress-62-intermediate-symlink",
+              "gold_finding_id": null,
+              "severity": "high",
+              "actionable": true,
+              "source_backed": true,
+              "line_anchored": true,
+              "has_fix_direction": true,
+              "has_regression_test_direction": true
+            },
+            {
+              "finding_id": "coderabbit-egress-62-positive-regression",
+              "gold_finding_id": null,
+              "severity": "medium",
+              "actionable": true,
+              "source_backed": true,
+              "line_anchored": true,
+              "has_fix_direction": true,
+              "has_regression_test_direction": true
+            }
+          ]
+        }
+      }
+    },
+    {
+      "case_id": "inkspan-pr-65",
+      "repository": "ContextualWisdomLab/inkspan",
+      "pull_request_number": 65,
+      "head_match": false,
+      "diff_size_bucket": "large",
+      "primary_language": "typescript",
+      "gold_findings": [],
+      "reviewers": {
+        "opencode": {
+          "triggered_attempts": 2,
+          "completed_attempts": 2,
+          "rate_limited_attempts": 0,
+          "infrastructure_only_reviews": 2,
+          "duplicate_reviews": 1,
+          "findings": []
+        },
+        "coderabbit": {
+          "triggered_attempts": 1,
+          "completed_attempts": 1,
+          "rate_limited_attempts": 0,
+          "infrastructure_only_reviews": 0,
+          "duplicate_reviews": 0,
+          "findings": [
+            {
+              "finding_id": "coderabbit-inkspan-65-doctoring-dates",
+              "gold_finding_id": null,
+              "severity": "low",
+              "actionable": true,
+              "source_backed": true,
+              "line_anchored": true,
+              "has_fix_direction": true,
+              "has_regression_test_direction": true
+            },
+            {
+              "finding_id": "coderabbit-inkspan-65-security-doc-sync",
+              "gold_finding_id": null,
+              "severity": "medium",
+              "actionable": true,
+              "source_backed": true,
+              "line_anchored": true,
+              "has_fix_direction": true,
+              "has_regression_test_direction": true
+            },
+            {
+              "finding_id": "coderabbit-inkspan-65-memory-claim",
+              "gold_finding_id": null,
+              "severity": "medium",
+              "actionable": true,
+              "source_backed": true,
+              "line_anchored": true,
+              "has_fix_direction": true,
+              "has_regression_test_direction": true
+            },
+            {
+              "finding_id": "coderabbit-inkspan-65-plan-count",
+              "gold_finding_id": null,
+              "severity": "low",
+              "actionable": true,
+              "source_backed": true,
+              "line_anchored": true,
+              "has_fix_direction": true,
+              "has_regression_test_direction": true
+            },
+            {
+              "finding_id": "coderabbit-inkspan-65-word-fixture-budget",
+              "gold_finding_id": null,
+              "severity": "high",
+              "actionable": true,
+              "source_backed": true,
+              "line_anchored": true,
+              "has_fix_direction": true,
+              "has_regression_test_direction": true
+            }
+          ]
+        }
+      }
+    }
+  ]
+}