Status
contract + direction
Conceptual layer
④ Decision
Repo layer
L4 knowledge-reasoning
Source
architecture section 8, section 3.6.4
Research
../../research/plant-efficiency-exploration-2026-09/20-oe-knowledge-retrieval.md
Normative
00-kernel.md — retrieved text never evaluates gates, invents ₹, or writes equipment.
Siblings
05-context-engineering.md · 06-memory.md · 11-models-and-seams.md · 13-improvement.md

Purpose#

Plant memory answers what worked here. Detectors answer what is happening now. The OE knowledge corpusVersioned literature + sparse method ontology; retrieved as advisory rows only answers: for this class of situation, what methods does industrial practice use to drive efficiency?

Without it, dual-family models lean on weights and thin playbook bullets. That is not enough when Stamped wants excellence across energy, cost, time, continuity, and exceptions.


Decision#

#TopicDecision
1GapFirst-class OE knowledge corpus — separate from Hindsight, case library, and PSM
2v1 retrievalSparse method ontology + vector retrieval (Vector-First or light Parallel Hybrid). Graph is methods↔scenarios↔signals, not plant topology
3PartitionAll hits → advisory ledger rows only
4AuthorityAdvisory. Case library wins outcomes. Calculator owns ₹. Code owns constraints
5Plant pathNo Agentic GraphRAG; no unbounded research loops
6OfflineCouncil / corpus curation may use heavier GraphRAG patterns to propose playbook bullets
7PinsCorpus version + embedding model in release lockfile

Rejected: dumping books into the system prompt; using PSMPlant Situation Model as the literature graph; letting chunks override hard stops; inventing rupees from a sourcebook example figure.

Would change this: Pilot evidence that vector-only (no method ontology) is enough; or that Graph-First is worth the latency for method selection.


Place in the stack#

OE corpus advisory

Context partitions

Advisory sources

OE corpus
vector + sparse graph

Playbook bullets
lockfile

Case library

Advisory partition

Dual-family draft

Hard gates unchanged

How to read it.

  1. OE corpus, playbooks, and case library feed only the advisory partition — never measured rows.
  2. Dual-family drafting may cite advisory row ids; uncited claims drop.
  3. Hard gates, constraint evaluator, and calculator path are unchanged.

Build now: Tier A ingest + vector index. Later: Parallel Hybrid default; Ask Adaptive Router (section 10 scope cut).

View Mermaid source
flowchart TB
    %% house-style: oe-corpus-advisory
    subgraph src["Advisory sources"]
        direction LR
        corp["OE corpus<br/>vector + sparse graph"]
        play["Playbook bullets<br/>lockfile"]
        case["Case library"]
    end
    subgraph part["Context partitions"]
        direction LR
        led["Advisory partition"]
        fam["Dual-family draft"]
    end
    gate{{"Hard gates unchanged"}}
    corp & play & case --> led --> fam --> gate

    classDef govc fill:#fff4d6,stroke:#c99a2e,color:#000
    classDef agentc fill:#e8f0ff,stroke:#5b7bd5,color:#000
    classDef loopc fill:#eef7ee,stroke:#4f9a4f,color:#000
    class gate govc

Ingest (summary)#

Detail and license notes: research 20.

TierExamplesv1?
A — publicDOE sourcebooks (air, pumps, fans, motors, steam, process heat); DOE tip sheets; EPA Lean & Energy toolkit; public IPMVP / ISO 50001 summariesYes
B — licensed / curatedFactory Physics, TOC, Roser papers → curated method cards; full text only with rightsSelective
C — plant-privateSOPs, treasure hunts, OEM manuals — owner-gated, plant-partitionedLater
OutMarketing % claims; OT-write instructions; chat logs; raw seriesNever

Runtime use#

  1. Code classifies situation → ScenarioClass (from family / pattern / scanner).
  2. Allowlisted mission (oe_methods_for_scenario, …) retrieves top-K chunks + method nodes.
  3. Rows appended to advisory partition with provenance=doc_id#chunk and corpus version.
  4. Models may cite those row ids. Uncited claims drop.
  5. Hard gates, constraint evaluator, calculator path unchanged.

Seam id (catalog add): oe_retrieval_mission — options from mission registry; disagreement → registry default (routing class). Jev may later replace mission choice, not the store.


Change class#

SurfaceClass
New public PDF in Tier AData (ingest job + lockfile corpus pin)
New method node / edgePlug-in / registry
New retrieval missionRegistry + replay
Letting corpus set money or hard gatesForbidden (kernel)

v1 slice vs later#

v1Later
Tier A ingest; vector index; hand-curated sparse ontology for Pilot scenariosBroader Tier B/C; Parallel Hybrid default; Ask Adaptive Router
Investigative lane + Ask retrieve OE knowledgeCertified fast path may pull a tiny pinned method card only
Offline council proposes playbook bullets from corpus + casesSame
Agentic GraphRAGOffline curation only

Page history: last 2 changes
  1. 2026-10-07 docs(technical): rewrite l4 21-30, glossary and README; reconcile architecture gaps e7fead7
  2. 2026-09-25 docs(l4): agentic decision architecture, ADRs, and production hardness 8275e7c

Diagram

100%

Search the architecture