Skip to content

PointSav Documentation

The engineering library for the PointSav platform — operating systems and services for regulated businesses that own their data, their AI, and their record-keeping outright. Where the monorepo holds the code, this wiki holds the reasoning: architecture, services, security, and the governance commitments that bind future development.

service-slm graph store rebuild

← All revisions

e4af38b1 · PointSav Digital Systems ·

fix(compliance): correct SYS-ADR-07/Ring-3 imprecision across 4 articles (EN+ES) — 'no runtime call reaches Ring 3' overstated the real rule; corrected to 'Ring 3 never writes, always via proposal + human approval + Ring 2 write path' per architecture/three-ring-architecture.md's own fuller definition. nightly-datagraph-rebuild.md's 'no AI inference/not model-generated' claims were flatly wrong (verified against service-extraction's real code and service-content's Doorman call) — corrected to describe AI-assisted extraction gated by human approval. service-slm-graph-store-migration.md's stale automated-write description corrected with the 2026-07-18 human-approval-gate fix, plus an unrelated short_description/body metadata mismatch (false LadybugDB-to-SQLite migration claim) fixed. Operator-confirmed resolution after joint investigation — not a real compliance breach, just documentation that predated the approval-gate fix.

View the full record as of this revision →

@@ -10,13 +10,21 @@ status: pre-build
audience: vendor-public
bcsc_class: current-fact
language_protocol: PROSE-TOPIC
last_edited: 2026-05-25
last_edited: 2026-08-01
editor: pointsav-engineering
paired_with: service-slm-graph-store-migration.es.md
short_description: "service-slm migrated its graph store from LadybugDB to SQLite for fleet nodes and runs a nightly DataGraph rebuild feeding inference context injection."
short_description: "service-slm's graph store runs a nightly DataGraph rebuild — grammar-constrained entity extraction via Doorman producing proposals, gated by human approval before any write — feeding inference context injection."
cites: []
---

**Correction (2026-08-01):** the short_description on this article previously claimed a
"migration from LadybugDB to SQLite" that the body below never described or supported —
a metadata/body mismatch, now corrected. The graph store described throughout this
article is LadybugDB; no SQLite migration is described here. Separately, the write flow
described below (the rebuild script calling `graph/mutate` directly) predates a human-
approval gate added to the DataGraph write path and proven end-to-end 2026-07-18 — see
the "Current status" correction near the end of this article.

The [[service-slm]] graph store is a live property graph of named business entities extracted nightly from an operator's data corpus — the entity layer that [[service-content]] uses to inject structured business context into every inference request without sending proprietary data to an external model. The graph is stored in LadybugDB and rebuilt on a nightly schedule by the DataGraph rebuild script, which runs as Phase 1 of the Elastic Compute nightly window before the model-training phase claims the GPU.

Each night, the DataGraph rebuild script processes the operator data
@@ -115,6 +123,19 @@ criterion before the DataGraph pattern is extended to larger operational
contexts. That gate has not yet been met; the rebuild pipeline is in its
initial operational period.

**Correction (2026-08-01):** the `graph/mutate` write step described above
predates a real correctness fix. Per the platform's [[architecture/three-ring-architecture|Three-Ring
Architecture]] rule, Ring 3 (Doorman/AI) output is always a proposal, never a
direct write — every accepted extraction proposal must pass a human approval
checkpoint before a Ring 2 write path commits it. That gate was built and
proven end-to-end on real data 2026-07-18 (extract → proposal → validate →
operator-approve → dry-run → confirmed commit). This article's routing-parity
description above (the rebuild calling the same two REST endpoints any
operator could call) is still accurate as far as it goes, but should now be
read as ending at a proposal, not an unconditioned write — the "no fabricated
commands"/"no defect hidden" integration-test framing holds, with the human
checkpoint as an explicit additional stage between extraction and mutation.

## See also

- [[elastic-compute-lora-training-pipeline]] — Phase 2 of the same nightly window (LoRA adapter training)
Important Information

Corporate structure. PointSav Digital Systems ("PointSav") is currently a trade name of Woodfine Capital Projects Inc. ("Woodfine"), planned to become a wholly-owned Woodfine subsidiary upon incorporation. PointSav does not itself offer, sell, or solicit any security. Any securities offering associated with Woodfine's real-property direct-hold solutions is made exclusively by Woodfine, and only by means of the applicable Private Placement Memorandum.

No investment advice. This wiki's content is provided for engineering, operational, research, and development purposes. Nothing on this wiki constitutes investment advice or a solicitation to invest in any Woodfine partnership or direct-hold solution.

Intellectual property. The PointSav name, trade name, wordmark, and marks, together with all current and future PointSav- and Totebox-branded products, services, and offerings — and the software, source code, documentation, design system, and all related materials — are proprietary to Woodfine and its affiliates, except for components identified as open source. No rights are granted except as expressly set out in a written license or agreement. The full trademark notice appears in the footer of every page on this site.

Open source components. Portions of the platform are made available under permissive open-source licenses identified in the accompanying repository. Use of those components is governed by their respective license terms.

No warranty; informational use. Content on this wiki is provided for general informational purposes only and does not constitute a representation, warranty, or commitment with respect to product functionality, availability, pricing, or roadmap. Some articles describe planned or intended features, capabilities, and milestones — language such as "planned," "intended," "targeted," "may," and "expected" marks this forward-looking content, which is subject to change and does not constitute a commitment regarding future performance.

Confidentiality. Where an article describes an operational or deployment detail that is not intended for public disclosure, that article is not published on this wiki. Content here is general-purpose engineering documentation, not customer-specific configuration.

Jurisdiction. Woodfine Capital Projects Inc. is organized in British Columbia, Canada. References to the Sovereign Data Foundation on this wiki describe a planned or intended initiative only, not a current equity holder or active governance body.

Changes to this notice. PointSav may update this notice from time to time; the version posted on this page governs.

Not a filing system. This wiki is not a securities filing system, an electronic disclosure repository, or a substitute for SEDAR+ or any other regulatory filing system. Formal securities filings are made through the applicable regulatory filing system, not through this wiki.

Full disclaimer. This notice supplements, and does not replace, the full Disclaimers article. In the event of any conflict, the full Disclaimers article governs.

Read the full disclaimer →