fix(strix): align fallback model contract

.github/workflows/opencode-review-dispatch.yml

@@ -4058,8 +4058,8 @@ jobs:
                   "apiKey": "{env:OPENAI_API_KEY}"
                 },
                 "models": {
-                  "gpt-5.6-luna": {
-                    "name": "OpenAI GPT-5.6 Luna (direct)",
+                  "gpt-5.4": {
+                    "name": "OpenAI GPT-5.4 (direct)",
                     "tool_call": true,
                     "reasoning": true,
                     "options": {
@@ -4471,17 +4471,13 @@ jobs:
           # or used for product/model improvement, so private repositories
           # include neither NIM nor anonymous free candidates and start at the
           # existing keyed fallback list: OpenCode Zen GPT-5.6 Terra, DeepSeek
-          # V3, the direct GPT-5.6 Luna slot, and pinned PAID
+          # V3, the direct GPT-5.4 slot, and pinned PAID
           # OpenRouter coder models (free-tier candidates hit the shared
           # free-models-per-day cap and hung for the full candidate timeout,
           # so the OpenRouter slots use cheap paid models billed against the
           # org's OpenRouter credits), then the full-size GPT-4.1 long-context
           # endpoint and provider-specific GPT/o3 fallbacks.
-          # The direct-OpenAI slot runs GPT-5.6 Luna: the newest family's
-          # cost-efficient tier, cheaper than the legacy gpt-5 it replaced
-          # ($1/$6 vs $1.25/$10 per 1M tokens) so the org OpenAI budget
-          # stretches further between top-ups.
-          OPENCODE_MODEL_CANDIDATES: "${{ needs.validate-pr-metadata.outputs.is_private == 'false' && 'nvidia-nim/nvidia/llama-3.3-nemotron-super-49b-v1.5 nvidia-nim/nvidia/llama-3.1-nemotron-ultra-253b-v1 nvidia-nim/nvidia/nemotron-3-super-120b-a12b nvidia-nim/nvidia/nemotron-3-ultra-550b-a55b nvidia-nim/meta/llama-3.3-70b-instruct nvidia-nim/deepseek-ai/deepseek-v4-pro nvidia-nim/mistralai/codestral-22b-instruct-v0.1 opencode-free/nemotron-3-ultra-free opencode-free/deepseek-v4-flash-free opencode-free/north-mini-code-free opencode-free/laguna-s-2.1-free opencode-free/ling-3.0-flash-free opencode-free/big-pickle opencode-free/mimo-v2.5-free opencode-free/hy3-free opencode-free/minimax-m3-free opencode-free/glm-5-free opencode-free/kimi-k2.5-free opencode-free/qwen3.6-plus-free ' || '' }}opencode/gpt-5.6-terra github-models/deepseek/deepseek-v3-0324 openai/gpt-5.6-luna openrouter/deepseek/deepseek-v3.2 openrouter/qwen/qwen3-coder github-models/openai/gpt-4.1 github-models/openai/gpt-5 github-models/openai/gpt-5-chat github-models/openai/o3 github-models/deepseek/deepseek-r1-0528 github-models/deepseek/deepseek-r1"
+          OPENCODE_MODEL_CANDIDATES: "${{ needs.validate-pr-metadata.outputs.is_private == 'false' && 'nvidia-nim/nvidia/llama-3.3-nemotron-super-49b-v1.5 nvidia-nim/nvidia/llama-3.1-nemotron-ultra-253b-v1 nvidia-nim/nvidia/nemotron-3-super-120b-a12b nvidia-nim/nvidia/nemotron-3-ultra-550b-a55b nvidia-nim/meta/llama-3.3-70b-instruct nvidia-nim/deepseek-ai/deepseek-v4-pro nvidia-nim/mistralai/codestral-22b-instruct-v0.1 opencode-free/nemotron-3-ultra-free opencode-free/deepseek-v4-flash-free opencode-free/north-mini-code-free opencode-free/laguna-s-2.1-free opencode-free/ling-3.0-flash-free opencode-free/big-pickle opencode-free/mimo-v2.5-free opencode-free/hy3-free opencode-free/minimax-m3-free opencode-free/glm-5-free opencode-free/kimi-k2.5-free opencode-free/qwen3.6-plus-free ' || '' }}opencode/gpt-5.6-terra github-models/deepseek/deepseek-v3-0324 openai/gpt-5.4 openrouter/deepseek/deepseek-v3.2 openrouter/qwen/qwen3-coder github-models/openai/gpt-4.1 github-models/openai/gpt-5 github-models/openai/gpt-5-chat github-models/openai/o3 github-models/deepseek/deepseek-r1-0528 github-models/deepseek/deepseek-r1"
           # One attempt per model, then fall through to the next model. Retrying
           # the SAME model 5x let a rate-limited/hung leader consume the whole
           # step, so the pool never reached a healthy fallback model.

CHANGELOG.md

@@ -70,6 +70,12 @@ Semantic Versioning where the repository publishes a release.
 
 ### Fixed
 
+- Reconciled the Strix required-workflow smoke contract and the privileged
+  OpenCode model pool with the current `gpt-5.4` direct-OpenAI fallback after
+  `gpt-5.6-luna` was retired. This prevents every consumer repository's
+  required Strix check from failing on a stale central assertion or selecting a
+  nonexistent direct model.
+
 - Publish only the sanitized cumulative Strix report tree, avoiding a later
   copy of relative scanner output that could reintroduce known internal warning
   text into uploaded security evidence.

docs/adr/0002-product-technical-gap-baseline.md

@@ -7,3 +7,15 @@
 - Ownership: .github owns control-plane evidence; naruon and product repositories own product behavior and consumer smoke.
 - Figma File ID: N/A. This repository has no customer UI. A UI-owning repository must replace N/A with its real Figma File ID before a UI PR is accepted and must provide Storybook and design-token evidence.
 - Consequence: The document is an operational snapshot, not a merge authorization or substitute for protected GitHub review. Hourly agents must re-collect exact head SHAs, reviews, threads, and required Checks before merge. Papers/standards live in `docs/doctoring/product-technical-gap-baseline.md` and must remain consistent with this ADR.
+
+## Amendment: central Strix fallback contract (2026-08-25)
+
+The current `main` workflow (`a724582`) intentionally replaced the unavailable
+direct-OpenAI `gpt-5.6-luna` fallback with `gpt-5.4`, but the required-workflow
+smoke script still asserted the retired model. The privileged OpenCode model
+pool also retained the retired candidate while its contract tests had already
+moved to `gpt-5.4`. This mismatch failed consumer Strix checks, including
+ContextualWisdomLab/disksage#247, before any target-repository security
+analysis ran. The workflow, smoke contract, model-pool configuration, and
+regression tests now share `gpt-5.4`; the change does not weaken provider
+failure or vulnerability fail-closed behavior.

docs/doctoring/strix-nvidia-nim-not-found-fallback.md

@@ -64,6 +64,15 @@ current NVIDIA documentation whenever a provider returns a catalog 404. This
 change does not treat arbitrary provider errors as success and does not weaken
 Strix severity, changed-file attribution, or independent approval requirements.
 
+## Current fallback contract (2026-08-25)
+
+The direct-OpenAI fallback is `gpt-5.4`. The retired `gpt-5.6-luna` identifier
+must not appear in the executable workflow, required smoke contract, or model
+pool. A central workflow update without its smoke and model-pool assertions is
+invalid because every consumer repository would fail before its own scan. The
+contract is verified by `scripts/ci/strix_required_workflow_smoke.sh` and the
+focused `test_strix_quick_gate.sh` case; provider failures remain non-passing.
+
 ## References
 
 Fielding, R., Nottingham, M., & Reschke, J. (2022). *HTTP semantics* (RFC

docs/product-technical-gap-baseline.md

@@ -204,6 +204,18 @@ flowchart LR
 - Open count is 98. No additional `.github` PR merged this pass.
 
 
+## 2026-08-25 central Strix fallback contract recheck
+
+- `main` at `a724582a0768129d481385070bf8f05b2620dd2c` changed the direct-OpenAI
+  fallback to `gpt-5.4`, but the required-workflow smoke script still required
+  the retired `gpt-5.6-luna` string. The privileged OpenCode model pool also
+  retained the retired candidate while its contract tests expected `gpt-5.4`.
+- This exact mismatch caused consumer Strix checks to fail before scanning the
+  target repository; it was observed on ContextualWisdomLab/disksage#247 at
+  exact head `a9c868a6e9c8d68a9c6ea6de381e188740b8f5db`. The focused repair keeps
+  provider errors and vulnerability findings fail-closed and only aligns the
+  executable model and its assertions.
+
 ## 5. 실행 루프와 고객의 다음 행동
 
 각 hourly pass는 아래 순서를 유지한다.

scripts/ci/emit_opencode_failed_check_fallback_findings.sh

@@ -956,7 +956,7 @@ extract_strix_failed_check_block "$EVIDENCE_FILE" "$strix_evidence_file"
 
 emit_known_missing_string_finding \
 	"$EVIDENCE_FILE" \
-	"steps.target_visibility.outputs.is_private == 'false' && 'nvidia_nim/nvidia/nemotron-3-ultra-550b-a55b' || 'gpt-5.6-luna'" \
+	"steps.target_visibility.outputs.is_private == 'false' && 'nvidia_nim/nvidia/nemotron-3-ultra-550b-a55b' || 'gpt-5.4'" \
 	"Strix public scans must default to NVIDIA NIM while private scans retain the contracted provider" \
 	".github/workflows/strix.yml" \
 	"scripts/ci/test_strix_quick_gate.sh"

tests/test_opencode_agent_contract.py

@@ -186,7 +186,7 @@ def test_opencode_model_pool_sets_high_effort_for_capable_candidates():
         ["opencode-free", "qwen3.6-plus-free"],
         ["opencode", "gpt-5.6-terra"],
         ["github-models", "deepseek/deepseek-v3-0324"],
-        ["openai", "gpt-5.6-luna"],
+        ["openai", "gpt-5.4"],
         ["openrouter", "deepseek/deepseek-v3.2"],
         ["openrouter", "qwen/qwen3-coder"],
         ["github-models", "openai/gpt-4.1"],
@@ -197,7 +197,7 @@ def test_opencode_model_pool_sets_high_effort_for_capable_candidates():
         ["github-models", "deepseek/deepseek-r1"],
     ]
     assert zen_models == ["gpt-5.6-terra"]
-    assert direct_openai_models == ["gpt-5.6-luna"]
+    assert direct_openai_models == ["gpt-5.4"]
     assert openrouter_models == [
         "deepseek/deepseek-v3.2",
         "qwen/qwen3-coder",
@@ -1740,7 +1740,7 @@ def test_workflow_provisions_sandbox_tool_and_reviewer_agent():
     assert (
         "opencode/gpt-5.6-terra "
         "github-models/deepseek/deepseek-v3-0324 "
-        "openai/gpt-5.6-luna "
+        "openai/gpt-5.4 "
         "openrouter/deepseek/deepseek-v3.2 "
         "openrouter/qwen/qwen3-coder "
         "github-models/openai/gpt-4.1 "
@@ -1887,7 +1887,7 @@ def test_workflow_provisions_sandbox_tool_and_reviewer_agent():
     )
     assert (
         "github-models/deepseek/deepseek-v3-0324 "
-        "openai/gpt-5.6-luna "
+        "openai/gpt-5.4 "
         "openrouter/deepseek/deepseek-v3.2 "
         "openrouter/qwen/qwen3-coder "
         "github-models/openai/gpt-4.1 "

tests/test_pr_review_autofix_nvidia_nim_contract.py

@@ -19,7 +19,7 @@
 DOCTORING_RECORD = Path("docs/doctoring/hourly-nvidia-nim-autofix.md")
 CHANGELOG = Path("CHANGELOG.md")
 REVIEW_DISPATCH_WORKFLOW = Path(".github/workflows/opencode-review-dispatch.yml")
-REVIEW_DISPATCH_BLOB_SHA = "dd65d90e10e5040562b501ade1a40f89572f0984"
+REVIEW_DISPATCH_BLOB_SHA = "ed3f7b44f9afdd6ab295426e5d0440aeca6bdfb5"
 
 
 def _workflow_text(path: Path) -> str:

tests/test_required_workflow_queue_contract.py

@@ -505,7 +505,7 @@ def test_nvidia_nim_defaults_preserve_existing_fallbacks_without_secret(
     assert strix.returncode == 0, strix.stderr
     assert {
         "provider_mode=openai_direct",
-        "strix_model=gpt-5.6-luna",
+        "strix_model=gpt-5.4",
     } <= set(strix_output.read_text().splitlines())
     assert (
         "STRIX_MODEL: ${{ steps.gate.outputs.strix_model }}"

tests/test_strix_nvidia_nim_not_found_fallback.py

@@ -192,7 +192,7 @@ def test_workflow_uses_available_free_first_nvidia_plan(self) -> None:
         workflow = STRIX_WORKFLOW.read_text(encoding="utf-8")
         default_expression = (
             "steps.target_visibility.outputs.is_private == 'false' && "
-            f"'{DEFAULT_NVIDIA_MODEL}' || 'gpt-5.6-luna'"
+            f"'{DEFAULT_NVIDIA_MODEL}' || 'gpt-5.4'"
         )
         self.assertIn(default_expression, workflow)
         self.assertIn(
@@ -202,7 +202,7 @@ def test_workflow_uses_available_free_first_nvidia_plan(self) -> None:
         )
         self.assertIn(
             "steps.gate.outputs.provider_mode == 'nvidia_nim' && "
-            f"'{FREE_NVIDIA_FALLBACK} openai-direct/gpt-5.6-luna'",
+            f"'{FREE_NVIDIA_FALLBACK} openai-direct/gpt-5.4'",
             workflow,
         )