Seed taxonomy as SMB bootstrap
Maintained by PointSav Digital Systems. Each revision is content-addressed by its commit hash.
- editorial(substrate): fix seed-taxonomy-as-smb-bootstrap's substantial mischaracterizations (Track-B) — confirmed real archetypes.csv has 11 entries (not 'five to seven'), listed as strategic/operational/physical-realization roles with signature+healing_trigger fields the article omitted; confirmed chart_of_accounts.csv is genuinely a personnel-role taxonomy (Director/Officer/Counsel/etc under sub-domains like Personal/Compliance/Finance), not 'industry-specific business profiles mirroring revenue and expense categories' as claimed — a fundamental mischaracterization; confirmed the entire 'gravity_keywords mechanism' section was fabricated — zero automated keyword-match-counting classifier exists anywhere (service-extraction has no archetype/gravity references at all); the one real consumer, Verification Surveyor (surveyor.py), shows only archetype names to a human who picks manually, gravity_keywords never read programmatically; kept exact Theme content generic per public-repo-safety-review.md given real themes.csv holds genuinely sensitive tenant-specific strategic content; register-clean EN+ES
- fix(index-group): backfill index_group across 11 categories migrated to Index Topic (330 files)
- feat(media-knowledge-documentation): add content_type field to all TOPIC + GUIDE files
- enrich wikilinks: documentation substrate body (EN+ES) batch 4 of 4
- editorial: Phase A2+A3 — title normalisation (169 files) + wikilink backslash fix (20 files)
- 2j: substrate EN + missed patterns/applications Bloomberg sweep (batch 6)
- docs(2g): corpus-wide Doctrine claim vocabulary removal (227 files)
- docs(2g): corpus-wide mechanical quality pass
- merge: integrate origin/main (taxonomy ratification + category-balance + schema-scrub commits) before project-knowledge Stage 6 promote
- Taxonomy ratification 2026-05-09: split architecture/ into substrate/+patterns/+architecture/; retire company/+help/