Releases · March 2026 · W11

v0.7.9 (2026-03-15)

Other

• Release 0.7.9
• Add —db and —afs-root flags to calibrate.py for live instance support

2 commits: 0 features, 0 fixes

---

v0.7.8 (2026-03-15)

Other

• Release 0.7.8
• Fix build order: lineage before reports (report_data_quality needs lineage_pipeline_metrics)
• Release 0.7.7
• Add pack-specific thesis to Domain tab (millesime winery thesis)
• Release 0.7.6
• Simplify Docker build: KLS-only (tenant config now lives inside DuckDB)

6 commits: 0 features, 0 fixes

---

v0.7.7 (2026-03-15)

Other

• Release 0.7.7
• Add pack-specific thesis to Domain tab (millesime winery thesis)

2 commits: 0 features, 0 fixes

---

v0.7.6 (2026-03-15)

Other

• Release 0.7.6

1 commits: 0 features, 0 fixes

---

v0.7.5 (2026-03-15)

Other

• Release 0.7.5
• Fix CI workflow: move platform filter from job-level if to step-level check

2 commits: 0 features, 0 fixes

---

v0.7.4 (2026-03-15)

Other

• Release 0.7.4

1 commits: 0 features, 0 fixes

---

v0.7.3 (2026-03-15)

Other

• Release 0.7.3
• Fix release script: don’t git-add RELEASE_NOTES.md (gitignored build artifact)
• more docs.
• Explorer: DB-first metadata reader, About page Domain/Mold tabs, jumpbar indexing
• Fix pack resolution bugs: stale bundled packs, dbt manifest labels, Docker AFS staging
• Fix millesime calibration: defect overlap, wine_id in lab CSVs, age-dependent sell patterns
• Increase synthetic data volume ~3.4x
• Fix traceability_gap defect injection: skip ALL cellar operations, not just RACK/FILL
• Merge remote-tracking branch ‘origin/docker-deployment’
• Deprecate numetrix auditreport.py: migrate all 5 report types to YAML
• more todos
• Fix calibration display for composite tenant IDs; skip DMG in release —now
• Fix PDF report rendering: flatten i18n labels, resolve field keys, add missing section titles
• Add self-contained Docker build for Tier 0 demo deployment
• Merge remote-tracking branch ‘origin/main’ into feature/mobile-explorer
• Add domain packs success story; untrack RELEASE_NOTES.md (build artifact)
• Enable calibration for millesime: find defect manifest via DLZ fallback
• Add CSV schema validation system (L0/L1/L2)
• Merge remote-tracking branch ‘origin/main’ into feature/mobile-explorer
• Release flags, better build errors, millesime fixes
• Add commune and exposition columns to millesime parcels pipeline
• Fix atomic build: use subdirectory with same filename for catalog name
• Fix KLS build temp file naming — keep .duckdb extension for catalog name
• Expand mobile spike into full Explorer UI with DuckDB-WASM

24 commits: 0 features, 0 fixes

---

v0.7.1 (2026-03-15)

Other

• Release 0.7.1
• release notes
• Rewrite CLI help text for all commands to reflect current capabilities
• Build KLS to temp file + atomic rename to avoid DuckDB lock conflicts
• Update release notes and add generic report pipeline design doc
• Fix jinflow status in deployed mode — show live tenants, not bundled pack
• Add three-party drift detection: jinflow status —tenant
• Add three-way sync to jinflow update — detect pack↔tenant conflicts
• Implement declarative calibration system (all 3 phases)
• Add pack-specific domain glossary and calibration design

10 commits: 0 features, 0 fixes

---

v0.7.0 (2026-03-15)

Other

• Release 0.7.0
• more todos
• Add generic PDF report renderer (reportrender.py)
• Add four millesime reports: vintage quality, customer portfolio, cellar inventory, appellation performance
• Add millesime taxonomies, raw→build passthrough, and gitignore alignment
• Add timestamp (H:M) to jinflow git log output

6 commits: 0 features, 0 fixes

---

v0.6.4 (2026-03-14)

Other

• Release 0.6.4
• Add two-level case_type (type+subtype), billing report breakdowns, and jinflow git log
• Remove standard_price validation; add backfill probe, billing report, and year-boundary SMEbit
• Bundle changelog in deployed binary for jinflow version —changelog
• Fix layout server to check new tenant/ path before legacy tenants/{tid}/
• more todos
• Fix _find_dbt_project_name to prefer pack name from manifest; promote 2 artifacts
• Add source_systems as pack property; lineage + entity YAMLs for millesime
• Add Data Landscape & Risk Report for millesime pack
• Enrich : 8x data volume with business complexity
• Fix multi-pack support in validators/compilers; add millesime probes and diagnosis
• Rewrite cheatsheets: CLI, justfile, Explorer, setup, and overview index
• Add T keyboard shortcut to navigate to tenant picker
• Fix Esc key not working in PDF viewer modal
• Extract PDF reports to disk after audit report generation
• Make report type list dynamic in Explorer reports page
• Add incremental make design doc and update TODOs
• Gitignore auto-generated registries and pipeline graph in tenant AFS
• Add ship recipe with benchmarks; fix tfoot alignment in report CSS
• Fix calamine row offset, update sync, and catalogue PDF visibility
• Add total rows to catalogue population report tables
• Remove all legacy tenants/{id}/ fallbacks from pack scripts
• Fix csvschemacheck.py for tenant/ consolidation (new → legacy fallback)
• Register DRG lookup as browseable dimension in Explorer

24 commits: 0 features, 0 fixes

---

v0.6.3 (2026-03-13)

Other

• Release 0.6.3
• Add missing _ok() output helper used by auto-commit

2 commits: 0 features, 0 fixes

---

v0.6.2 (2026-03-13)

Other

• Release 0.6.2
• Fix enrich_csvs.py for tenant/ consolidation (new → legacy fallback)
• Add just log recipe for quick commit history
• Add DRG column docs to source gold.yml schema
• Wire drg_lookup seed into gold_cases with bilingual descriptions
• Add jinflow git status/push commands and auto-commit on update

6 commits: 0 features, 0 fixes

---

v0.6.1 (2026-03-13)

Other

• Release 0.6.1
• Add catalogue population PDF audit report
• Add reports model directory to dbt_project.yml
• do it:-)
• Consolidate tenants/{tenant_id}/ → tenant/ across codebase
• Fix nav sidebar highlighting overlapping /report and /reports
• Add catalogue population model and report

7 commits: 0 features, 0 fixes

---

v0.6.0 (2026-03-13)

Other

• Release 0.6.0
• Add jinflow reset command and tenant/seeds/ convention
• Fix jinflow update to sync .sql files; remove stale result reference
• Remove catalogue_type from Bronze; extend jinflow update for default/wildcard tenant
• Add jinflow update CLI command and article_catalogues lineage definition
• a study for data modelling.
• Add CatalogueType entity and derive article_type from veska_account at Silver
• Remove enrich step from millesime pipeline and engine-level redirect
• Formalize source_type: seed in lineage validator and compiler
• Fix make-all subprocess invocation using sys.argv[0]
• more todos.
• Add jinflow make-all CLI command and service mandate generation
• Fix dimensions 500 for composite tenant IDs; add rebuild-all script
• Make pipeline graph generator pack-agnostic
• Add tenant branding to pack splash screen
• Fix noisy Bun segfault crash dump on Explorer shutdown
• Add jinflow branding: logo, pack splash screen, tenant-agnostic SMEbits
• Add testing concept: 4-tier framework from black-box to AI content validation
• Fix Explorer queries for composite tenant IDs; add release —now

19 commits: 0 features, 0 fixes

---

v0.5.3 (2026-03-13)

Other

• Release 0.5.3
• Make Explorer and build pipeline pack-agnostic; enrich millesime
• Add Explorer generic report renderer for YAML-defined reports
• Wire report build pipeline and fix io_coefficient probe SQL errors
• Add generic report framework: YAML-defined reports compiled to dbt SQL

5 commits: 0 features, 0 fixes

---

v0.5.2 (2026-03-12)

Other

• Release 0.5.2
• Add jinflow pick — interactive tenant picker

2 commits: 0 features, 0 fixes

---

v0.5.1 (2026-03-12)

Other

• Release 0.5.1
• Add cross-pack tenant discovery and contract-driven Explorer
• Wire CLI commands into evolve REPL shell

3 commits: 0 features, 0 fixes

---

v0.5.0 (2026-03-12)

Other

• Release 0.5.0
• Move csvschemacheck.py from engine scripts to numetrix pack
• Replace —dlz with —dlzroot and —dlztenant flags
• Add millesime pack source models directory
• Add wine_id to millesime lab_analyses pipeline
• Add HarvestTrend entity and glossary for millesime pack
• Fix millesime hypothesis YAMLs (short_code, categories)
• Rewrite millesime probe YAMLs to conform to DSL
• Make all 8 YAML compilers pack-generic
• inspired; todo
• Data-driven taxonomy, layered DLZ resolution, and XLSX float fix

11 commits: 0 features, 0 fixes

---

v0.4.16 (2026-03-12)

Other

• Release 0.4.16
• one more todo
• Fix multi-pack init and StepResult.elapsed typo

3 commits: 0 features, 0 fixes

---

v0.4.15 (2026-03-12)

Other

• Release 0.4.15
• Add jinflow generate command, make AFS pack-generic
• Add DuckDB-WASM mobile spike for Android feasibility test
• Refactor Millesime generator to use —output-dir argument
• Add health endpoint, Dockerfile healthcheck, and Fly.io config
• Update cheatsheet and release paths

6 commits: 0 features, 0 fixes

---

v0.4.14 (2026-03-12)

Other

• Release 0.4.14
• Scale up Millesime synthetic data and output to DLZ
• Add standalone jinflow extract command, remove evolve ingest
• Add changelog, timestamps, verbose build, fix probe-builder paths, consolidate releases
• Fix release summary: look for zips in correct location
• Fix wiki sync: point to jinflow.wiki instead of numetrix-1st.wiki
• Add report: XLSX extraction calamine migration findings
• Add Millésime domain pack: winery operations analytics for the
• Switch XLSX extraction from openpyxl to calamine (Rust, ~5x faster)

9 commits: 0 features, 0 fixes

---

v0.4.13 (2026-03-11)

Other

• Release 0.4.13
• Extend raw extraction to E3-E4 and E5, move all interpretation to Bronze
• Move releases/ out of deploy root to prevent DMG bloat

3 commits: 0 features, 0 fixes

---

v0.4.12 (2026-03-11)

Other

• Release 0.4.12
• Split E1 extraction from Bronze normalization, rename script
• Timestamp build log filenames (build-YYYYMMDD-HHMMSS.log)
• Write build.log next to KLS after jinflow make
• Add —extract flag to jinflow make for XLSX to CSV extraction
• Fix severity_case_sql: handle above_money key in severity rules
• Show artifact sizes in release summary

7 commits: 0 features, 0 fixes

---

v0.4.11 (2026-03-11)

Other

• Release 0.4.11
• Fix exponential DMG growth — write DMG outside deploy root
• Fix double-zip: let upload-artifact handle zipping

3 commits: 0 features, 0 fixes

---

v0.4.10 (2026-03-11)

Other

• Release 0.4.10
• Fix Windows read_text() cp1252 decode errors

2 commits: 0 features, 0 fixes

---

v0.4.9 (2026-03-11)

Other

• Release 0.4.9
• Fix Windows cp1252 encoding errors
• Add three “Introducing jinflow” slide decks

3 commits: 0 features, 0 fixes

---

v0.4.8 (2026-03-11)

Other

• Release 0.4.8
• Zip CI artifacts flat — no nested subfolders

2 commits: 0 features, 0 fixes

---

v0.4.7 (2026-03-11)

Other

• Release 0.4.7
• Add release recipes with semver bump and update README

2 commits: 0 features, 0 fixes

---

v0.4.6 (2026-03-11)

Other

• Release 0.4.6
• Fix release.sh: allow untracked files, fix clean check
• Fix CRLF line endings in release.sh
• Add scripts/release.sh: full release pipeline

4 commits: 0 features, 0 fixes

---

v0.4.5 (2026-03-11)

Other

• Fix GitHub Actions release notes: grep exit code and shallow clone
• Add jinflow release-notes command, bump to 0.4.5

2 commits: 0 features, 0 fixes

---

v0.4.4 (2026-03-11)

Other

• Add release notes: generated from commits, shown in CLI and Explorer
• Add Linux build to GitHub Actions workflow
• Fix Windows encoding errors and shrink deploy binary by ~1.2 GB
• Update pyproject.toml description

4 commits: 0 features, 0 fixes

---

v0.4.3 (2026-03-11)

Other

• Add nuMetrix setup guide with Mermaid diagrams, bump to 0.4.3
• Fix UTF-8 encoding in _update_raw CSV row counter on Windows

2 commits: 0 features, 0 fixes

---

v0.4.2 (2026-03-11)

Other

• Force UTF-8 file encoding on Windows, show DLZ in us, bump to 0.4.2
• Show DLZ in jinflow us, fix _resolve_setting priority

2 commits: 0 features, 0 fixes

---

v0.4.1 (2026-03-11)

Other

• Fix GitHub Actions: install full deps including dbt for PyInstaller
• Add version number to GitHub Actions artifact filename
• jinflow CLI including make, evolve, explore, status, us.
• Suppress requests charset detection warning in frozen binary
• Bundle requests package in PyInstaller to fix charset detection warning

5 commits: 0 features, 0 fixes

---

v0.4.0 (2026-03-11)

Other

• Bump version to 0.4.0
• Bundle charset_normalizer in PyInstaller to fix requests warning
• Bundle domain pack in PyInstaller binary for jinflow init
• Fix config priority: user config always beats repo jinflow.yml
• Fix live root resolution for frozen binary, improve error message
• Add jinflow cheat sheet, fix WeasyPrint fontTools bundling in spec
• Bundle WeasyPrint dependencies in PyInstaller, add report templates to pack
• Fix auditreport.py to resolve KLS path dynamically, fix init test AFS path
• Bundle weasyprint in deployed binary, fix native lib discovery on macOS
• Skip audit reports gracefully when weasyprint is not installed
• Fix test: system prompt now mentions write_artifact instead of READ-ONLY
• Add write/delete artifact tools to evolve REPL, fix pipeline graph AFS resolution
• Add pipeline graph and audit reports to make, show brand name in Explorer

13 commits: 0 features, 0 fixes

---

v0.2.0 (2026-03-10)

Other

• Remove test step from Windows build workflow
• Add GitHub Actions workflow for Windows desktop build
• Add /make command to evolve REPL, bundle anthropic in deployed binary
• Fix evolve AFS resolution for deployed binary
• Fix explore AFS resolution for deployed binary, auto-find free port
• Track jinflow.spec in git (PyInstaller build config)
• Self-contained deployed binary: embedded Python, dbt, and script execution
• Add secrets.yml, model config, and evolve KLS auto-discovery
• Add user-level config, build-desktop flags (—skip, —dmg), register slow marker
• Fix 5 test failures: isolate repl fixtures from pack artifacts
• Run tests before build in build-desktop.sh
• more todos!
• Add jinflow clone command, Explorer DB info button, and fix stale tests
• Add jinflow us and jinflow ls commands for config visibility and tenant listing
• Add full environment status and default tenant resolution
• Fix CRLF line endings in build scripts
• Add build scripts and jinflow explore —docker
• Add report framework design: YAML-defined reports compiled to KLS
• Replace hardcoded nuMetrix branding in report pages with dynamic brand config
• Add architecture docs, Millesime pack design, and gitignore cleanup
• Add cloud make/evolve security analysis and KLS delivery strategy
• Fix Docker build: exclude build artifacts from Tailwind v4 oxide scanner
• Reset drops tenant schemas instead of deleting the entire KLS file
• Delete existing KLS on reset, auto-detect live AFS from cwd
• Enforce pack read-only: make/snapshot always target live instances
• Remove compiled dbt models from pack — models/ is 100% generated
• Add Dockerfile for cloud Explorer deployment
• Fix KLS naming: use underscore separator (dbt-duckdb splitext compatibility)
• Wire KLS to live root, rename NUMETRIX_* env vars to JINFLOW_*, migrate system DB to DuckDB
• Rename .jinflow.yml to jinflow.yml, add DLZ integration and source system discovery
• more todos!
• Move hand-written SQL from dbt output to pack source directories
• Add pipeline.yml hook system for extensible make phases
• Copy all platform models during init (fix missing ref warnings)
• Fix CSV validation for live instances via —csv-dir override
• Add —update-raw and —input/—output for live instance pipeline
• Move enrich_csvs.py from engine to domain pack
• Add live root architecture, deploy root, and configurable dbt path macros
• just dtm/dtc to report from every docs folder.
• some principal data quality thoughts.
• the beginning of millesime
• ecosystem docs
• tauri dependencies
• more todos…
• Add KLS deploy root, snapshots, tenant qualification, and CLI improvements
• Fix Tauri bundle copy path for jinflow branding
• Add Tauri desktop build setup to setup.sh
• Add jinflow-intake data separation, taxonomy generation in enrichment
• final setup activities.
• Replace hardcoded nuMetrix branding with dynamic .jinflow.yml brand config
• Update jinflow evolve AI model to claude-sonnet-4-6
• Remove synthetic tenants, update CLAUDE.md for pack structure
• Fix gitignore for old dbt/evidence paths and update compiled registries
• some compilation artefacts.
• more to do
• Add concept paper: Provenance, Custody, and Retention
• Fix 6 test failures: probe counts, —check timestamps, and round-trip tests
• Phase 3: Move domain scripts, docs, and evidence under pack
• Phase 1+2: Move domain content to packs/numetrix/
• Phase 0: Abstract domain path resolution through AFS pack discovery
• Update entity_registry compiled_at timestamps
• Add jinflow CLI implementation reference document
• Add comprehensive tests for evolve REPL and brand config
• Add jinflow explore command and improve REPL prompts
• some learnings.
• Add AF Convention tab, artifact decoder, ELK layout engine, and pipeline persistence
• Add git worklog script — daily session tracker from commit timestamps
• Update registry compiled_at timestamps from rebuild, remove stale calendar_events
• Add “Inspired By” generator: extract statistical profiles, synthesize unlimited hospitals
• Add “Inspired By” concept paper for statistical fingerprint synthesis
• Add white-label branding via .jinflow.yml brand config
• Make jinflow an installable Python package
• Sort just pick recipe list alphabetically (a-z)
• Show KLS and SIS in evolve REPL banner
• Add jinflow evolve REPL with dual-mode local/AI exploration
• Add What’s New page, artifact timestamps, and return-highlight UX
• Fix dbt 1.11 accepted_values deprecation in gold.yml
• Switch docs-by-mtime/ctime recipes from filesystem to git dates
• Add docs-by-mtime and docs-by-ctime recipes to justfile
• more and more
• Add concept papers: KLS registry completeness and What’s New
• Update todo list and pin pnpm packageManager
• Add cross-site price consistency and catalogue completeness probes
• Fix back-link history loop: pop stack instead of pushing
• Add SVG favicon for Explorer
• Fix backspace navigation: use window.history.back()
• Add JumpBar tab to About page with keyboard shortcuts and search guide
• Add backspace navigation and state-preserving back links
• Add back navigation to dimension pages
• Add dimension name search to JumpBar
• Add KLS hot-swap: auto-invalidate registry caches on DB rebuild
• Add catalogue count reconciliation probe and entity registration
• Add article_catalogues entity across Bronze/Silver/Gold/Platform
• Add article_catalogues CSV extraction for catalogue membership (m:n)
• Add jinflow make, evolve ingest, portable app concept, and CLI tests
• Update Claude Code permission settings
• Add dbt engine to AFS init: selective copy of models, macros, seeds
• Add jinflow CLI and —afs flag to all compilers and validators
• Add ArtifactStore API, jinflow init, and AFS-aware CSV intake
• Add AFS (Artifact Store) concept document with branches-as-scenarios
• Revamp About page Architecture tab with jinflow mindset
• Update jinflow explore: desktop or web, not desktop-only
• Add jinflow mindset presentation (Make, Explore, Evolve + AFS/KLS/SIS)
• Fix About page glossary showing English for all locales
• Fix adjusted exposure returning 0: ambiguous column reference in DuckDB
• Fix 373M CHF overcounting on overview: use entity-deduplicated adjusted exposure
• Introduce AFS/KLS/SIS store acronyms in jinflow concept and glossary
• Add jinflow toolset vision & concept document
• Update docs references from ddd-glossary.md to terminology.md, minor wording fixes
• Add terminology.yaml compiler with glossary.json + terminology.md generation
• Replace fictional with Claude (nuMetrix Analytical Engine)
• Replace fictional SME names with Claude (nuMetrix Analytical Engine)
• Rename BitBundle to Use Case in UI labels, add tenant terminology concept
• Rename provider/curator from to in SMEbits and BitBundles
• Internationalize the About page (DE/FR/EN) with 114 i18n keys per locale
• Add Executive Summary and Catalog Inventory report pages, update demo slidedeck
• Recompile SMEbit and BitBundle registries with latest YAML definitions
• Filter tenant picker to tenants with DuckDB data, auto-redirect on single tenant
• Delete dev.duckdb before reta/retall for a clean zeta-only database
• Add JumpBar filter mode (: prefix) for in-page table filtering
• Add 45 SMEbits and 13 BitBundles across all tenants, backfill why fields
• Add catalog inventory KPI slides to demo walkthrough
• Add 4 inventory KPI probes + catalog inventory assessment
• Fix SMEbit name in demo walkthrough to match actual subject
• Add just reta/retall shortcuts for hospital_zeta rebuilds
• Add demo walkthrough presentation and fix stale test counts
• Add people search to JumpBar
• Unify SMEbit/BitBundle provider to
• Fix locale reactivity in Explorer data-driven text
• Add three-layer architecture concept (core → erp → numetrix)
• Add project-level Claude Code permissions
• Add root node_modules to gitignore and deduplicate entries
• Add knowledge-driven workflow doc and presentation
• Enrich validity probes with purchasing activity classification
• Add SMEbits and BitBundles to JumpBar search and navigation
• Add auto video demo scaffolding (Playwright + GitHub Actions)
• Add hypothesis, diagnosis, and SMEbit for catalog maintenance gap
• Add sentinel date probes and catalog health assessment
• Add supplier/channel FK validation and fix channel code mismatch
• Add diagnosis: no dual sourcing policy (structural, confidence 0.90)
• Add hypothesis: supplier concentration risk (confirmed on hospital_zeta)
• Add BitBundle: Supplier Data Quality at
• Add Piota duplicate SMEbit (Level 1) and single-source dependency probe
• Add SMEbit: E6 supplier coverage gap (Level 0, structural)
• Add system database foundation (SQLite) for operational state
• Add hospital_zeta supplier & article activity analysis report
• Trim Tauri build to Windows-only (MSI)
• Ingest E6 Société Code into material_sourcing, relax XOR to OR
• Update todo: add product, cloud, and video items
• Replace article_suppliers with material_sourcing + procurement_channels
• Add CSV schema check (Step 1c) to catch column mismatches before dbt build
• Update About page: add SMEbits/BitBundles, drop hard-coded model counts
• Add BitBundle system: curated narrative groupings of SMEbits
• Add material procurement model Phase 1: Gold contract + models
• Replace null with ”*” for cross-tenant SMEbit scope, scope fictional bit to alpha
• Add orphan material SMEbit, tenant scope filtering, fix rebuild Phase 1 selector
• Add SMEbit Explorer pages: list, detail, sidebar nav, tri-lingual i18n
• Rename skills → commands (correct Claude Code terminology)
• Add Claude Code skills and design principles doc
• Add SMEbit system documentation to CLAUDE.md
• Add SMEbit pipeline: contract, validator, compiler, dbt models, platform layer
• Move four presentation decks to docs/presentations/
• Update About page: add diagnosis system, updated counts, new glossary entries
• Add DSL naming conventions presentation deck (28 slides)
• Add introduction and data engineers presentation decks
• Add software engineers presentation deck (39 slides)
• WIP: design principles doc, enrich_csvs script, synthetic data updates
• Atomize hypothesis and diagnosis compilers
• Move 3 informational docs (generic_design, roadmap, strategic_directions) to narrative/
• Graduate 7 implemented design docs to architecture/
• Add BitBundles concept doc and refine SMEbits
• Reorganize docs/ into 7 category folders
• more todos
• Add data_source and is_backfilled lineage columns across all dimension entities
• Explorer: convention-driven auto-filters for dimension list pages
• Add press release for entity registry migration
• Entity registry: replace all hardcoded Explorer maps with registry-driven lookups
• concepts and todos
• Jumpbar: dynamic entity search, add taxonomy results
• Add concept docs for anonymisation and three-layer split
• Add SMEbits concept doc — first-class subject matter expert knowledge
• more
• ignore build artefacts
• more todos.
• housekeeping
• Remove stale .app before copying to dist/
• Add live elapsed timer to jin-all.sh
• Add jin-all.sh: parallel rebuild + Tauri build script
• Fix provider CC collision in H4 taxonomy, add member captions, copy db to dist
• Ground ou_structure taxonomy in H4 service mandate routing data
• Add authorization & journal concept — capability RBAC + hash-chained audit
• Add OU/CC analysis docs, inferred data audit, ERP org structure paper; fix tests and jumpbar
• Pipeline graph: add switchable shape mode (Layer vs Kind)
• Update cloud concept to v2.0 — processing/exploration separation
• Pipeline graph: light theme, visible-only re-layout, detail panel scroll fix
• Serve probe YAMLs from DuckDB pipeline graph, add tenant switching to jumpbar search
• more to do.
• Add desktop builds doc, fix justfile DMG copy error
• Fix pnpm version detection in CI — point to explorer/package.json
• Use database in-place instead of copying to ~/.numetrix
• Use dynamic port allocation for Node.js sidecar
• Add week bucket to probe builder and YAML parser
• Add Open Database menu for switching databases at runtime
• Skip DMG bundling — use app/msi/deb targets only
• Fix file dialog deadlock — use rfd before Tauri event loop
• Update pipeline graph
• Switch all probes from monthly to weekly time buckets
• Add GitHub Actions cross-platform Tauri builds and database-at-launch flow
• Remove translated/variant presentations, consolidate insights
• Add conceptual introduction presentation — 38-slide overview of nuMetrix
• Add I/O Imbalance presentation and hypothesis short codes on overview
• Fix all 70 svelte-check warnings and 1 type error
• Store pipeline graph as per-tenant audit artifact in DuckDB
• Add PascalCase labels, larger fonts, outline mode, and dark gray canvas to pipeline graph
• Replace probe-centric overview with hypothesis-centric executive summary
• Add source code inspection to pipeline graph — slide-over panel with syntax highlighting
• Add qty_at_risk to findings contract, probes, hypotheses, and Explorer
• Update todo — mark completed items, add new tasks
• Add Executive Brief narrative to hypothesis presentation page
• Update todo and add hypothesis presentation plan doc
• Enhance hypothesis presentation — fix trend charts, add calendar overlay, resolve type errors
• Fix pipeline graph search — dimming, match navigation, concept doc
• Add Jump Bar — keyboard-driven command palette for Explorer
• Move hypothesis short_code from prefix badge to trailing parenthetical
• Add weekly I/O coefficient, hypothesis short codes, and presentation page
• Remove text outline from pipeline graph node labels
• Add audit readiness assessment slide deck and roadmap
• Update pnpm-lock.yaml
• Move probe builder and playground to System section in sidebar
• Add probe playground and builder with roundtrip text layer
• Add interactive pipeline graph — generator script + Explorer integration
• Add jinflow.io presentation — framework, domain packs, AI as SME
• Convert Beyond the Numbers to auto-forward with progress bar
• Add “Beyond the Numbers” presentation — lives, careers, and the patient behind every finding
• Extend Three Compilers presentation with Reverse Compiler section
• Add Three Compilers presentation — probe, hypothesis, diagnosis + forward compiler ambition
• Convert treatment presentation to sidebar+thumbnail style
• Add 72h sprint report (22–25 Feb 2026)
• Add ward-level I/O coefficient, service mandates entity, and reference seed tables
• Fix CC-level I/O coefficient — location→CC mapping, substr bug fix
• Sharpen I/O coefficient — overlap period filter, population classification, compiler WHERE support
• Update defect manifests after rebuild
• Add subcategory depth + English labels to MIGEL taxonomy
• Add Treatment layer concept doc and diagnosis-to-treatment presentation
• Add requirements.txt, missing lookup tables analysis, and MIGEL taxonomy docs
• Update todo.md — mark completed items, add remaining priorities
• Add confidence weighting, entity dedup, and rebuild diagnosis step
• Add strategic directions document — five paths evaluated, Diagnosis + Temporal chosen
• Add Diagnosis layer — rule-based root cause explanations for confirmed hypotheses
• Add risk_tier to probes — tiered financial risk in reports and Explorer
• reports and slidedecks
• Fix PDF reports for Tauri builds — blob-based viewer + download
• Update convert_zeta_xlsx.py for H4 columns and 5-digit CC codes
• Add Analytics Readiness report type to auditreport.py
• Add domain-portability to docs index (For Decision-Makers)
• Add GitHub Wiki auto-sync from docs/
• Add proberegistry.py — auto-generate probe_registry.sql from YAML + glossary
• Add probe_billing_below_internal_price to assessment_material_health
• Add enrichment probe DSL type and probe_billing_below_internal_price
• some git VIP
• Add reconciliation probe DSL type for dual fact-table joins
• the learnings.
• keep learnings in repo!
• Format monetary amounts, percentages, and dates in generic data tables
• Add extended presentation, tri-lingual versions, adaptive timing, and slide counter
• Serve all reports and calibration data from DuckDB, not filesystem
• Store calibration & audit report data in DuckDB
• Add domain portability analysis document
• Documentation overhaul — update narrative docs, add index and looping presentation
• Unify build pipeline: just rebuild delegates to rebuild.sh
• Fix weasyprint library discovery on macOS/Homebrew
• Add assessment design improvements — multi-field/compound severity rules, health score, composability
• Fix Tauri desktop app launch — shell config and duckdb binding
• Update todo checklist, remove stale calibration reports
• Add force-directed graph view on taxonomy cardinality page
• Fix Tauri app startup: remove stale shell scope, patch duckdb binding
• Replace default Tauri icon with nuMetrix branded logo
• Add member-level drill-down on taxonomy cardinality page
• Add clickable fanout bars as filters on taxonomy cardinality page
• Update calibration report timestamps
• Update todo.md with probe config and PDF report items
• Add test suites, bootstrap workflow, and cardinality enhancements
• Add taxonomy cardinality cross-check — dbt model + Explorer page
• Add timestamps to reports and calibration, manifests for all synthetic tenants
• Track app-icon.png so tauri icon works on fresh clones
• Add prerequisite checks to just tauri for sidecar and icons
• Update todo.md
• Add calibration harness — defect manifest, calibrate.py, Explorer integration
• Add audit report generation — PDF reports, Explorer page, rebuild integration
• Add weighted assessments and graduated hypothesis scoring
• Add just tauri recipe for desktop app builds
• Package Explorer as desktop app with Tauri v2
• Update todo.md
• Add self-documentation lineage layer — YAML definitions, compiler, runtime metrics
• Add member search to taxonomy view — ILIKE on dimension_member_id
• Update about page and explain-like.md for hypothesis layer
• Add Glossary tab to about page with domain vocabulary
• Add taxonomy tree/table view toggle, search filtering, and green color tokens
• Add full-width tenant banner, remove sidebar tenant switcher, persist tenant in localStorage
• Add phantom cases hypothesis, improve verdict badges, disable partial parsing
• Add hypothesis system — business questions evaluated against probe evidence
• Add “What is this?” about page to Explorer + update explain-like.md
• Extend Evidence dashboard + add cloud deployment concept
• Add generic_design concept doc and update todo
• Fix ATC taxonomy for SAP and Navision tenants
• Add MIGEL taxonomy builder with per-tenant BAG reference
• ou-cc-analysis
• migel!
• Fix evidence JSON, add ATC taxonomy, i18n fallback, declarative related facts
• Replace static DIMENSIONS array with data-driven schema discovery
• Add 11 V4.0 probes, 2 assessments, DSL extensions, and Explorer updates
• Widen E1+E4 ingestion to V4.0 fields
• Add assessments tab and overview cards to explorer
• Merge compiler output into single pass, one line per probe
• Add assessment probe type and rename all probes with type prefix
• Add probe-reset step to drop stale findings before rebuild
• Add probe-run-all recipe for stage 3 across all tenants
• Remove money_at_risk from silver probes, use count-only severity
• Add data_layer column to probe registry, show silver badge in explorer
• Add silver contract and extend probe DSL to compile all 10 probes from YAML
• Extend probe DSL with duplicate type, relax mandatory_item
• Add 3 new probes across data_quality and compliance categories
• Move year selector from hero header into Activity section
• Add probe catalog page, split YAML per probe, category-grouped overview
• Update todo and add claude statements doc
• Add first data_quality probe, fix CC mapping, rework article_suppliers
• Add Assessment phase and is_valid semantics to probe system
• Add probe_category to registry, clean up overview page
• Remove source_system as required CLI parameter — auto-resolved from tenant.yml
• Add duplicate billing probe — flags identical billing events
• Add what_this_is.md — high-level vision and approach document
• Fix DuckDB catalog mismatch: copy as explorer/dev.duckdb
• Merge feat/explorer-app into main
• Dynamic probe system: registry, discovery, templates, and tests
• Add comprehensive pipeline overview for non-technical audience
• Fix explorer DuckDB not actually opening read-only
• Update project todo list
• Add I/O coefficient trend detection model and probe
• Fix dates rendering as [object Object] in explorer
• Add just rebuild recipe and explorer pnpm lockfile
• Document all 68 dbt models with descriptions and column definitions
• Add setup script and README for first-time repo setup
• Merge pull request #1 from jazzisnow/feat/explorer-app
• Update justfile and welcome.html
• Add I/O coefficient models and probe
• Fix date display, add compact formatting, fix probe fan-out bug
• Housekeeping: auto-resolve source_system, supplier metrics, verbose rebuild
• Replace misleading total money-at-risk with per-probe breakdown
• Add hero header, year selector, and billing stats to Explorer
• Add monogram logo for hospital_zeta
• Add nuMetrix Explorer: customer-facing SvelteKit app
• Refine hospital_zeta dimensions: OU hierarchy, classifications, ATC taxonomy
• Add hospital_zeta to dashboard rebuild panel
• Add —inject flag for Mermaid ER diagram in welcome.html
• Add DuckDB → Mermaid ER diagram generator
• Add justfile CLI layer with tenant context and fzf integration
• Add dimension ingestion planning docs
• Add reference dimensions (H1, H2, E6) and extend material attributes
• Add execution timing and summary table to rebuild.sh
• Fix platform model cross-schema dependencies for clean builds
• Update welcome.html to reflect current project state
• Gitignore .claude/ directory and untrack settings.local.json
• Add dimension taxonomy models, zeta converter, and dev utilities
• Add per-tenant tenant.yml config files
• Rename vendor → source_system across codebase
• Add probe findings dashboard and platform union view
• Add probe DSL framework v1: YAML-defined diagnostics compiling to dbt SQL
• Clarify platform glossary entry in welcome page
• Add CSV format docs, welcome page, and dashboard server
• Update Dagster orchestration for hospital_delta and multi-tenant support
• Regenerate synthetic data for alpha, beta, and gamma tenants
• Update .gitignore: exclude hospital_delta data, logs, workspace files
• Add rebuild.sh script for tenant regeneration
• Add hospital_delta: large-scale tenant with advanced defect injection
• Add Navision vendor support with hospital_gamma tenant
• Add user guide and multi-level explainer docs
• Multi-tenant architecture with SAP vendor support and operational Dagster
• Add Evidence dashboard with revenue integrity, operations, and logistics pages
• Bootstrap nuMetrix hospital material flow lakehouse

379 commits: 0 features, 0 fixes