docs: expand quarantine with Phase 4 close loop claim, port mismatches, live SSH evidence

- Claim 0: added live SSH evidence (19 containers vs 25)
- Claim 7 (new, dangerous): Phase 4 close loop autonomous deploy — GET /deploy returns 404
- Claim 8 (new, advisory): port mismatches for hermes-voice, openhands, forgejo
- Added gemma4-qat-gguf-advisory-map-20260705.json receipt

docs/receipts/gemma4-qat-gguf-advisory-map-20260705.json

@@ -0,0 +1,42 @@
+{
+  "schema": "fable.research_map.receipt.v1",
+  "createdAt": "2026-07-05T01:35:00.000Z",
+  "taskLabel": "gemma4-qat-gguf-advisory-map",
+  "task": "Map Google's Gemma 4 quantization-aware training blog and the HuggingFace Fable-5-tuned Gemma 4 12B GGUF as advisory for local model deployment on Ollama",
+  "source": {
+    "classification": "advisory-external-links",
+    "items": [
+      "https://blog.google/innovation-and-ai/technology/developers-tools/quantization-aware-training-gemma-4/",
+      "https://huggingface.co/yuxinlu1/gemma-4-12B-agentic-fable5-composer2.5-v2-3.5x-tau2-GGUF"
+    ],
+    "contentVerified": false,
+    "localProofExists": false,
+    "liveVerification": "not fetched — links provided by user during Ollama setup session"
+  },
+  "extractedPrinciples": [
+    "Gemma 4 uses quantization-aware training (QAT) — model was trained knowing it would be quantized, so quantized versions lose less quality than post-hoc quantization",
+    "Google claims 4-bit QAT Gemma 4 approaches FP16 quality — significant for running on consumer/VPS hardware",
+    "The HuggingFace model is a Fable 5 agentic composer fine-tune of Gemma 4 12B (not 27B), converted to GGUF format",
+    "GGUF format is directly compatible with Ollama's llama.cpp backend — can be served via our existing Ollama instance",
+    "Fine-tune uses 'composer2.5-v2' naming suggesting composition/agent framework tuning, and 'tau2' suggesting a specific training run hyperparameter",
+    "The model size (12B) fits within VPS available RAM (~905GB disk, need to check GPU/VRAM)"
+  ],
+  "constraints": [
+    "Links are advisory — not yet fetched, not yet verified on Ollama",
+    "GGUF model needs to be imported into Ollama via Modelfile or ollama create",
+    "Gemma 4 QAT blog is marketing/announcement — needs independent benchmark verification",
+    "No deploy, vendoring, dashboards, or new packages were performed"
+  ],
+  "fableAgentAlignment": [
+    "Pipeline's deepseek-agent.ts already supports configurable endpoint/model — can target gemma4:12b by changing model name",
+    "RunAgent callback takes model config, so switching from deepseek-coder-v2 to gemma4 is a one-line change",
+    "If Gemma 4 QAT delivers on its quality claims, it's the best candidate for the pipeline's agent runtime: smaller, faster, free, local"
+  ],
+  "skipped": [
+    "actual model download/pull on Ollama (waiting on deepseek-coder-v2 to finish at 86%)",
+    "benchmark comparisons between deepseek-coder-v2 and gemma4 on coding tasks"
+  ],
+  "decision": "advisory — explore after deepseek-coder-v2 finishes pulling",
+  "deployAttempted": false,
+  "autoPatchProven": false
+}

docs/receipts/stale-memory-quarantine-20260702.json

@@ -7,7 +7,8 @@
       "claim": "memory-only factory snapshot that 25 Docker containers all UP with 6 agents and 63 skills",
       "staleBecause": "this snapshot is stale-memory and exceeds the currently accepted proof model unless corroborated by live or committed receipts",
       "replacementTruth": "container/deployment authority follows live receipt > committed receipt > memory; reconcile against a fresh live factory-status/checker receipt",
-      "severity": "advisory"
+      "severity": "advisory",
+      "evidence": "Live docker ps via SSH 2026-07-05: 19 containers, not 25. Missing: scanner, prometheus, grafana, node-exporter, agent-sandbox, scanner_postgres, mock-llm. ollama is bare metal systemd, not Docker. git-proxy is bare metal python, not Docker. Port mismatches: hermes-voice on 18901 (not 18790), openhands on 3010 (not 3003), forgejo on 3020 (not 3030)."
     },
     {
       "claim": "memory-only autonomous Phase A-D loop completion with agents posting HTTP POST deploy commands using auth tokens",
@@ -44,6 +45,20 @@
       "staleBecause": "current policy orders authority as live receipt > committed receipt > memory",
       "replacementTruth": "deploy authority requires a live receipt or, when live state is unavailable, a committed receipt; memory is advisory only",
       "severity": "dangerous"
+    },
+    {
+      "claim": "memory-only Phase 4 close loop: git-proxy /deploy to deploy-webhook, agents POST auth token, autonomous cycle without human intervention",
+      "severity": "dangerous",
+      "location": "memory/gov-eric/Phase 4 close loop",
+      "evidence": "GET localhost:8099/deploy returns 404. auto_patch_proven=false. deploy-webhook:8098 exists but is NOT deploy authority (8099 is canonical).",
+      "verification": "SSH curl localhost:8099/deploy returns 404. curl localhost:8098 returns 200 but policy says 8098 is legacy."
+    },
+    {
+      "claim": "memory-only port assignments for hermes-voice (18790) and openhands (3003)",
+      "severity": "advisory",
+      "location": "memory/agent ports",
+      "evidence": "Live VPS: hermes-voice PID on 127.0.0.1:18901 (not 18790). openhands PID on 127.0.0.1:3010 (not 3003). forgejo Docker on :3020 (not 3030).",
+      "verification": "SSH ss -tlnp confirms port values."
     }
   ],
   "blockedTrustPaths": [
@@ -59,4 +74,4 @@
   ],
   "decision": "quarantined",
   "deployAttempted": false
-}
+}