Integrate OCR 1.11.5 with private tool diagnostics

.github/workflows/ocr-compatibility.yml

@@ -6,7 +6,7 @@ on:
   workflow_dispatch:
     inputs:
       tag:
-        description: Optional exact stable upstream tag, for example v1.7.18
+        description: Optional exact stable upstream tag to qualify against the current contract
         required: false
         type: string
 

AGENTS.md

@@ -16,7 +16,7 @@ Use this file as the short repository map and startup workflow for Open Code Rev
 - `docs/engineering/toolkit_strategy.md` and `ROADMAP.md` - durable direction and outcome state.
 - `docs/engineering/project_principles.md` - cross-cutting engineering invariants and ownership boundaries; `docs/engineering/m5_context_contracts.md` owns the current context and evidence contracts.
 - `docs/review-decision-flow.md` - canonical detailed Mermaid map for review, diagnostics, receipts, DLP, publication, and later-action decisions; keep it synchronized with runtime and public contracts.
-- `docs/development.md` - implementation workflow, boundary checklists, and local validation.
+- `docs/development.md` - implementation workflow, boundary checklists, local validation, and [OCR qualification maintenance](docs/development.md#maintaining-ocr-qualification).
 - `docs/release.md` - release classification, authorization, publication, and archival lifecycle.
 - `docs/codex/TASKS_BACKLOG.md` - inactive work with activation conditions.
 - `docs/codex/AGENT_EXECUTION_PITFALLS.md` - incident catalogue for diagnosis, not an instruction source.

README.md

@@ -30,13 +30,12 @@ python -m venv .venv
 python -m pip install open-code-review-toolkit
 ```
 
-The toolkit does not bundle OCR. Download the platform asset for exact
-[Open Code Review 1.11.3](https://github.com/alibaba/open-code-review/releases/tag/v1.11.3),
-verify it before installation against the [compatibility manifest](compatibility/ocr-support.json),
-and place the verified binary on `PATH`. The current Linux amd64 digest is
-`9726204ac81baee153fd65b1ff357c380f73e9d8091c4a73c3c9fb541b5164cb`; Darwin arm64 is
-`515cd92ce300b62dea2bcaf53e910cdf57b681e1d58bbd0a9e8e49617bca52fa`. The manifest owns
-the other platform hashes.
+The toolkit does not bundle OCR. Use `recommended_version` in the
+[compatibility manifest](compatibility/ocr-support.json) for this toolkit revision
+to select the exact [upstream release](https://github.com/alibaba/open-code-review/releases).
+Download its platform asset, verify it against that release entry's asset SHA-256,
+and place the verified binary on `PATH`. Do not substitute the latest upstream
+release or use another toolkit revision's manifest.
 
 Verify the installation without contacting an LLM provider:
 
@@ -45,14 +44,14 @@ ocr --version
 ocr-ci --help
 ```
 
-`ocr --version` must report `open-code-review v1.11.3`; `ocr-ci --help` must exit
+`ocr --version` must report the manifest's exact `recommended_version`; `ocr-ci --help` must exit
 successfully. `ocr-ci preflight` is an operational configuration check, not the installation
 smoke test.
 
 The exact recommended OCR release and its verified asset checksums live in the [versioned compatibility manifest](compatibility/ocr-support.json). CI should pin that release and checksum before execution.
 The [versioned compatibility policy](docs/compatibility.md) records tested assets and evidence and describes the conservative Dependabot-like qualification workflow for later upstream releases.
 Review output defaults to English. `OCR_REVIEW_LANGUAGE` accepts another explicit language name when a project needs localized review output; for example, `OCR_REVIEW_LANGUAGE=Russian`.
-The current OCR 1.11.3 integration defaults `OCR_REVIEW_EFFORT` to `medium` for two review rounds. `low` and `high` are explicit one- and three-round alternatives; see the [configuration reference](docs/configuration.md#review-effort) for cost, budget, and precedence boundaries.
+The toolkit defaults `OCR_REVIEW_EFFORT` to `medium` for two review rounds. `low` and `high` are explicit one- and three-round alternatives; see the [configuration reference](docs/configuration.md#review-effort) for cost, budget, and precedence boundaries.
 
 Stable distributions are published to [PyPI](https://pypi.org/project/open-code-review-toolkit/) and mirrored as checksum-listed, provenance-attested assets in the corresponding [GitHub Release](https://github.com/xeonvs/open-code-review-toolkit/releases). Development snapshots are published only to TestPyPI.
 

changelog.d/176.bugfix.md

@@ -0,0 +1,3 @@
+Accept bounded private `failure_details.arguments` from OCR without treating otherwise valid failed-tool diagnostics as malformed. Arguments are omitted before normalized console output and publication DLP; they never enter finalized results, receipts, GitLab notes, or toolkit telemetry. Invalid diagnostics still preserve valid review findings and summary publication. Existing DLP checks on public findings, warnings, and suggestions and independent later-action restrictions remain in effect.
+
+Fail OCR qualification probes through the closed compatibility-status path when a review emits malformed JSON or invalid result, tool-call, failure-detail, or comment shapes. Private diagnostics remain available on stderr for CI troubleshooting without leaking into the public status artifact.

changelog.d/176.maintenance.md

@@ -0,0 +1 @@
+Qualify OCR 1.11.4 as the separately recorded predecessor to the final 1.11.5 target. Verify native comment-batch recovery and its repair warning through a deterministic gateway, while preserving finding fields, anchors, suggestions and rejection of suspect truncation. Upstream site and plugin-distribution changes do not add toolkit capabilities. See the separate diagnostic-parser fix and Objective-C++ Rules entries.

changelog.d/176.refactor.md

@@ -0,0 +1 @@
+Separate forward-only live OCR qualification from frozen historical evidence validation. Every candidate now runs the same current behavioral suite; legacy grouping/preview execution fallbacks are removed, promotion validates current contract evidence before writing, and generic promotion tests no longer track the latest runtime pin. Existing recorded evidence remains readable and unchanged. This changes repository qualification tooling, not the toolkit runtime API.

changelog.d/176.rules.md

@@ -0,0 +1 @@
+Route Objective-C++ `.mm` files through OCR's built-in Objective-C rules. Content-dependent MATLAB/Objective-C routing for `.m` remains qualified.

changelog.d/177.doc.md

@@ -0,0 +1,3 @@
+Document group-scoped concurrency and prompt ceilings, the separate per-file large-diff filter, and the diagnostic-argument discard step in the canonical review decision flow. Distinguish private OCR artifacts from published GitLab summaries and identify the built toolkit artifact required for pre-release qualification.
+
+Keep current operator guidance version-neutral and point installation and pin lookup to the same toolkit revision's compatibility manifest; preserve exact executable pins and historical qualification evidence.

changelog.d/177.maintenance.md

@@ -0,0 +1 @@
+Target checksum-verified OCR 1.11.5 as the only supported runtime for toolkit 0.9.1; install it directly without an intermediate OCR release. Separately qualify the 1.11.4 and 1.11.5 changes. Grouped-review naming now describes concurrency and prompt ceilings per group; per-file large-diff filtering, default medium effort, inherited completion cap, max-tools, result/manifest and receipt contracts remain unchanged. The OCR binary includes the grpc update; VS Code dependency updates and browser-only viewer fixed/ignored marks add no toolkit behavior or GitLab lifecycle authority.

compatibility/evidence/ocr-1.11.4.json

@@ -0,0 +1,257 @@
+{
+  "assets": [
+    {
+      "name": "opencodereview-darwin-amd64",
+      "sha256": "09a57a9022071c19f82cec4d793bb8d93d3b308e0207bdd6b9ad823cbb1e5d17",
+      "size": 57803552
+    },
+    {
+      "name": "opencodereview-darwin-arm64",
+      "sha256": "36a29bb525f59ad88f39f6fb4a8bd76a67df5ed95399c92f3d10143274c91b15",
+      "size": 54844930
+    },
+    {
+      "name": "opencodereview-linux-amd64",
+      "sha256": "7133465774aff6330fd8d25eecc57a28e5852bc55d1ae55fd00443ebc4c6b023",
+      "size": 56058018
+    },
+    {
+      "name": "opencodereview-linux-arm64",
+      "sha256": "981964ac4466f1ed20129ae46b63a31a929219c1228a7a94c345cb4c9b8155d3",
+      "size": 52822178
+    },
+    {
+      "name": "opencodereview-windows-amd64.exe",
+      "sha256": "e9300d32877a1512e9571a5776c9173c2f7eccee58d2f7ef4aa54011bd70b9fb",
+      "size": 57519616
+    },
+    {
+      "name": "opencodereview-windows-arm64.exe",
+      "sha256": "e08fc1d3896509b1589046f6180b92b921cd540c113faff60a98d4f79198450d",
+      "size": 53482496
+    },
+    {
+      "name": "sha256sum.txt",
+      "sha256": "2dccf4dccaabcb0be94e43766316acbf7c8ab6f7309001575ee9e0cd4862a273",
+      "size": 572
+    }
+  ],
+  "classification": "human-review-required",
+  "classification_reasons": [
+    "release notes contain a material or ambiguous compatibility signal"
+  ],
+  "comparison_version": "1.11.3",
+  "contracts": {
+    "comment_arguments_probe": {
+      "comments_per_batch": 2,
+      "failure_arguments_preserved": true,
+      "intact_cases": [
+        "array",
+        "serialized",
+        "repaired"
+      ],
+      "repair_warning": true,
+      "result": "passed",
+      "suspect_batch_rejected": true
+    },
+    "comment_thinking_probe": {
+      "additive_field_preserved": true,
+      "posting_exposes_thinking": false,
+      "result": "passed"
+    },
+    "completion_cap_probe": {
+      "explicit": 4096,
+      "inherited": 16384,
+      "result": "passed",
+      "wire_field": "max_completion_tokens"
+    },
+    "language_rule_probe": {
+      "excluded_extensions": [
+        ".svh"
+      ],
+      "extensions": [
+        ".cjs",
+        ".cxx",
+        ".hxx",
+        ".mjs",
+        ".mm",
+        ".pug",
+        ".sv",
+        ".v",
+        ".vh",
+        ".vhd",
+        ".vhdl"
+      ],
+      "m_routing": "matlab_and_objective_c",
+      "result": "passed",
+      "rule_source": "system_builtin",
+      "selected": 11
+    },
+    "numeric_cli_probe": {
+      "max_tokens_budget": {
+        "cases": {
+          "invalid_below": {
+            "effective": null,
+            "input": -1,
+            "outcome": "rejected"
+          },
+          "minimum": {
+            "effective": 1,
+            "input": 1,
+            "outcome": "accepted"
+          },
+          "omitted": {
+            "effective": "unlimited",
+            "input": null,
+            "outcome": "accepted"
+          },
+          "representative": {
+            "effective": 30000,
+            "input": 30000,
+            "outcome": "accepted"
+          },
+          "sentinel": {
+            "effective": "unlimited",
+            "input": 0,
+            "outcome": "accepted"
+          }
+        },
+        "maximum": null,
+        "owner": "ocr-cli"
+      },
+      "max_tools": {
+        "cases": {
+          "invalid_below": {
+            "effective": null,
+            "input": -1,
+            "outcome": "rejected"
+          },
+          "minimum": {
+            "effective": 100,
+            "input": 50,
+            "outcome": "accepted"
+          },
+          "minimum_minus_one": {
+            "effective": 100,
+            "input": 49,
+            "outcome": "normalized",
+            "reported_normalization": 50
+          },
+          "omitted": {
+            "effective": 100,
+            "input": null,
+            "outcome": "accepted"
+          },
+          "representative": {
+            "effective": 101,
+            "input": 101,
+            "outcome": "accepted"
+          },
+          "sentinel": {
+            "effective": 100,
+            "input": 0,
+            "outcome": "accepted"
+          }
+        },
+        "maximum": null,
+        "owner": "ocr-template-or-higher-cli",
+        "reported_minimum": 50
+      },
+      "result": "passed"
+    },
+    "optional_capabilities": [
+      "llm_result_identity",
+      "per_run_model_override",
+      "per_run_provider_override",
+      "review_effort",
+      "semantic_grouping"
+    ],
+    "preview_probe": {
+      "format": "json",
+      "path": "example.py",
+      "result": "passed",
+      "session_store_created": false
+    },
+    "required_review_flags": [
+      "--audience",
+      "--background-file",
+      "--effort",
+      "--format",
+      "--from",
+      "--max-tokens-budget",
+      "--max-tools",
+      "--preview",
+      "--rule",
+      "--to"
+    ],
+    "result_contract_probe": {
+      "additive_fields_allowed": true,
+      "comment_fields": [
+        "category",
+        "content",
+        "end_line",
+        "existing_code",
+        "path",
+        "severity",
+        "start_line",
+        "thinking"
+      ],
+      "manifest_schema": "ocr.run-manifest/v1",
+      "normalized_outcome": "clean",
+      "result": "passed"
+    },
+    "review_budget_probe": {
+      "budget": 30000,
+      "completed": 2,
+      "failed_budget": 1,
+      "grouping_requests": 0,
+      "grouping_strategy": "per_file",
+      "partial_findings_preserved": true,
+      "result": "passed",
+      "selected": 3
+    },
+    "semantic_grouping_probe": {
+      "default_effort": "medium",
+      "files": 4,
+      "filter_requests": 1,
+      "grouping_completion_cap": 16384,
+      "grouping_requests": 1,
+      "main_requests": 3,
+      "prior_finding_semantics": "filter_survivors_as_confirmed",
+      "recheck_instruction_requests": 3,
+      "result": "passed",
+      "review_rounds": 2
+    },
+    "small_change_grouping_probe": {
+      "grouping_requests": 0,
+      "high_churn": "per_file",
+      "low_churn": "bundle_all",
+      "result": "passed",
+      "single_file": "per_file",
+      "threshold_files": 4
+    },
+    "target_rule_selection_probe": {
+      "format": "json",
+      "from_to_unchanged": true,
+      "path": "synthetic-template.ocrfixture",
+      "result": "passed",
+      "source_exclusion": "unsupported_ext",
+      "target_selected": true
+    },
+    "version_probe": "passed"
+  },
+  "published_at": "2026-09-04T12:38:24Z",
+  "qualification_provenance": {
+    "asset_verification_run": "https://github.com/xeonvs/open-code-review-toolkit/actions/runs/33962853525",
+    "contract_binary_sha256": "36a29bb525f59ad88f39f6fb4a8bd76a67df5ed95399c92f3d10143274c91b15",
+    "contract_platform": "darwin-arm64"
+  },
+  "release_changes": "## 🚀 Features\n\n- feat(tool): include arguments in failure details (#1152)\n\n## 🐛 Bug Fixes\n\n- fix(rules): route Objective-C++ files to Objective-C rules (#1166)\n- fix(tool): repair serialized code_comment arguments instead of losing the batch (#1153)\n\n## Other Changes\n\n- Keep repeated documentation headings independently addressable (#1151)\n- ci: add plugin distribution contract guardrails (#1117)\n\n**Full Changelog**: https://github.com/alibaba/open-code-review/compare/v1.11.3...v1.11.4",
+  "release_notes_sha256": "81f1a6a8b329e688b2575f69b779dc9593e407763783ffbff47cdcacb86febff",
+  "result": "compatible",
+  "schema_version": 3,
+  "tag": "v1.11.4",
+  "tested_baseline_version": "1.11.3",
+  "upstream_repository": "alibaba/open-code-review",
+  "version": "1.11.4"
+}