ewoooc/docs/guides/ai_automation_mainline_work_items.md

AI Automation Mainline Work Items

Updated: 2026-07-10 Scope: EwoooC / MOMO Pro System AI automation, PixelRAG, MCP/RAG, production UX.

Source Of Truth

• Production readback stays authoritative after deploy: /health, container health, AI automation API readback, and smoke.
• Gitea is the only source-control truth. GitHub remains frozen and must not be used.
• Runtime truth must stay separate from source truth: a green test or registry does not mean MCP/RAG runtime is enabled.
• PixelRAG visual receipts are evidence, not formal price truth. They cannot write competitor_prices, competitor_price_history, or ai_insights without replay and promotion gates.

P0

Status	Work item	Evidence / next machine action
Completed	Multi-commerce PixelRAG visual evidence lane for momo, pchome, shopee_tw, coupang_tw, yahoo_shopping_tw, etmall_tw, friday_tw, rakuten_tw	API and CLI can emit expansion plans and marketplace manifests; PChome parser failures can queue manifests.
Completed	External MCP/RAG capability inventory absorbed into internal governance readback	/api/ai-automation/external-mcp-rag-integration and scripts/ops/report_external_mcp_rag_integration.py expose 9 capabilities, absorbed/unresolved counts, and runtime flags.
Completed	PixelRAG receipts to internal RAG candidate replay	/api/ai-automation/pixelrag-rag-candidate-replay and scripts/ops/report_pixelrag_rag_candidate_replay.py read visual capture receipts and platform probe worker receipts, split eligible / blocked / source-contract-fallback candidates, and expose combined OCR/VLM + source-contract next actions before identity matching and PromotionGate knowledge writes.
Completed	PixelRAG application portfolio and integration lanes	/api/ai-automation/pixelrag-application-portfolio and scripts/ops/report_pixelrag_application_portfolio.py expose commerce, RAG, UX, ops, marketing, and governance uses with priority, status, next machine action, and forbidden guardrails.
Completed	PixelRAG Ollama-first VLM replay worker	/api/ai-automation/pixelrag-vlm-replay-worker and scripts/ops/run_pixelrag_vlm_replay_worker.py dry-run or execute ready visual receipts against approved Ollama VLM routes, emit evidence-bound artifact receipts, and keep blocked pages out of product data.
Completed	PixelRAG VLM route readiness and auto model select	/api/ai-automation/pixelrag-vlm-route-readiness and scripts/ops/report_pixelrag_vlm_route_readiness.py read approved Ollama /api/tags, expose configured/candidate model readiness, and let execute mode avoid missing-model blind generate calls.
Completed	PixelRAG platform probe worker	/api/ai-automation/pixelrag-platform-probe-worker and scripts/ops/run_pixelrag_platform_probe_worker.py turn platform probe plans into dry-run/execute automation: Shopee interstitials go to public empty-context capture when runtime is available, missing Playwright auto-falls back to structured-source/backoff receipt, and Coupang 403/access denied goes to structured-source/backoff artifact receipts.
In progress	MCP/RAG runtime health in AI automation smoke	/api/ai-automation/smoke and /api/ai-automation/scheduled-health-summary include external MCP/RAG integration, PixelRAG RAG candidate replay, PixelRAG OCR/VLM replay contract, PixelRAG VLM route readiness, PixelRAG VLM replay worker, PixelRAG platform probe, and PixelRAG platform probe worker families.
In progress	Formal production deploy/readback discipline	Every mainline change must update version, push Gitea main/dev, deploy to 188 without touching momo-db, and read back /health plus new endpoints.

P1

Status	Work item	Evidence / next machine action
Completed	OCR/VLM replay contract for visual fields	/api/ai-automation/pixelrag-ocr-vlm-replay and scripts/ops/report_pixelrag_ocr_vlm_replay.py turn saved tiles into no-write field contracts, output schemas, validation rules, and Ollama-first worker actions.
Completed	Ollama-first OCR/VLM extraction worker	run_pixelrag_vlm_replay_worker.py --execute --write-receipt executes the replay contract against approved Ollama multimodal routes and emits confidence/evidence artifact receipts without writing formal price truth.
Not started	Ollama-first multimodal embedding benchmark	Verify local Qwen3-VL or equivalent visual embedding on GCP-A -> GCP-B -> 111 before any visual vector retrieval.
Not started	pgvector-compatible visual evidence metadata	Design metadata-first retrieval without FAISS in production unless ADR approves a different store.
In progress	Coupang platform probe / structured API strategy	Treat 403 as platform barrier; run_pixelrag_platform_probe_worker.py --execute --write-receipt now emits structured-source/backoff receipts without network, DB writes, or product-data promotion.
In progress	Marketplace source contracts beyond MOMO/PChome	Platform probe worker receipts now enter RAG candidate replay as source-contract fallback candidates; next machine action is to replay stable Shopee/Coupang/Yahoo/ETMall/friday/Rakuten source contracts only after provenance, rate-limit, public boundary, and replay evidence.
In progress	Professional product website UI/UX text reduction	Continue compact AI surfaces, visual QA, and user-facing copy guardrails; avoid engineering logs or work-session prose in the UI.

Always Enforced

• Manual review is exception-only for low-risk, machine-verifiable controlled apply.
• Break-glass still applies to secrets, destructive DB actions, force pushes, production provider switchovers, raw runtime volumes, and unverified core model replacement.
• AI automation smoke must expose health honestly: MCP_ROUTER_ENABLED=false or RAG_ENABLED=false is a runtime gap, not a source-code failure.