star-magic-program 0.395.0__tar.gz → 0.396.0__tar.gz

{star_magic_program-0.395.0 → star_magic_program-0.396.0}/CHANGELOG.md

@@ -7,6 +7,38 @@ Versioning: [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
 ---
 
+## v0.396.0 — 2026-08-24 — THE TWO-STREAM SHIP
+
+`uqff_downhole_simulator` v1.7.0–v1.9.0 — Daniel's two-stream architecture
+(closed theoretical stream vs live site stream, coordinated to find offset
+undervalued data) built COMPLETE:
+
+- **v1.7.0 the tool library** — cited catalog of 8 toolstring entries: quartz
+  twins (wrapping the GEOQ 177 specs), piezoresistive class (drift FORM cited
+  from ChampionX — exponential in temperature; coefficients disclosed as
+  representative fit), vibrating wire / thermocouple card / DTS fiber as
+  PARAMETERS_USER_SUPPLIED (refusing invented vendor numbers), and the G6
+  Modbus-RS485 surface interface declared as the ports target. ToolString +
+  station rating checks (catches a 177 °C gauge in the 217 °C kick zone).
+- **v1.8.0 the ports/plug-in layer** — READ-ONLY ingest to the normalized
+  LiveStream form. IMPLEMENTED: historian CSV (round-trip verified against the
+  v1.3.0 telemetry export to precision) and LAS 2.0 (CWLS standard; NULL→NaN;
+  wrapped mode refused, never mis-parsed). DECLARED (refusing until site
+  details): modbus_g6, witsml, opcua. register_port() for site plug-ins;
+  CLI `ingest`.
+- **v1.9.0 the reconciler** — per-station offset series classified IN_FAMILY /
+  CALIBRATION_OFFSET / DRIFT_CONSISTENT / TRANSIENTS / UNEXPLAINED, thresholds
+  disclosed in every report. Four-scenario validation per gate run: clean well
+  6/6 in-family; +50 psi bias recovered at 50.09; 2-yr drift inside the
+  spec-aware envelope; THE FIND — kick well vs linear-gradient assumption
+  flags 610/4,448/6,083 psi unexplained offsets. CLI `reconcile`.
+
+Also: v0.395.0 red-gate portability fix hardened (PORTABILITY GUARD self-scan).
+Gate 5,801 → 5,816. PAPER_2256 +3 appendices (6/7/8). Zero calculator physics
+changes; zero new dispatches.
+
+---
+
 ## v0.395.0 — 2026-08-23 — THE INSTRUMENT SHIP
 
 `uqff_downhole_simulator` v1.2.0–v1.6.0 (Daniel-directed extension list, complete):

{star_magic_program-0.395.0 → star_magic_program-0.396.0}/CITATION.cff

@@ -3,8 +3,8 @@ message: "If you use Star-Magic-Program (UQFF systematic rebuild) in your resear
 type: software
 title: "Star-Magic-Program: UQFF Systematic Rebuild"
 abbreviation: "UQFF"
-version: "0.395.0"
-date-released: "2026-08-23"
+version: "0.396.0"
+date-released: "2026-08-24"
 authors:
   - family-names: "Murphy"
     given-names: "Daniel T."
@@ -53,7 +53,7 @@ preferred-citation:
     - family-names: "Murphy"
       given-names: "Daniel T."
   year: 2026
-  version: "0.395.0"
+  version: "0.396.0"
   url: "https://github.com/Daniel8Murphy0007/Star-Magic-Program"
   notes: >-
     Cite both the software and the relevant whitepapers by PAPER_N

{star_magic_program-0.395.0/star_magic_program.egg-info → star_magic_program-0.396.0}/PKG-INFO

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: star-magic-program
-Version: 0.395.0
-Summary: Star-Magic UQFF calculator v0.395.0: THE INSTRUMENT SHIP. 2,252 dispatches (2,307 keys), gate 5,801/0. uqff_downhole_simulator v1.2.0-v1.6.0: service-life divergence curves, field-telemetry realism with scored QC, depth-sweep customer case studies, web-verified datasheet gauge specs (GEO PSI GEOQ 177 / Quartzdyne), MD/TVD wellbore deviation, multi-well batch runs, headless CLI. Suppression ratio 1.0324 invariant across geometry and baseline. Next: PAPER_2257.
+Version: 0.396.0
+Summary: Star-Magic UQFF calculator v0.396.0: THE TWO-STREAM SHIP. 2,252 dispatches (2,307 keys), gate 5,816/0. uqff_downhole_simulator v1.7.0-v1.9.0 completes Daniel's two-stream architecture: cited tool library (8 entries, no invented vendor data), read-only ports (historian CSV + LAS 2.0 implemented, live taps declared), and the two-stream reconciler - closed-stream predictions vs live data, offsets classified, undervalued streams flagged (kick-well find: 6,083 psi unexplained). Next: PAPER_2257.
 Author-email: "Daniel T. Murphy" <daniel.murphy00@enrgyone.com>
 License: AGPL-3.0-or-later OR LicenseRef-StarMagic-Commercial
 Project-URL: Homepage, https://github.com/Daniel8Murphy0007/Star-Magic-Program
@@ -34,17 +34,17 @@ Dynamic: license-file
 
 # Star-Magic-Program
 
-[![PyPI version](https://img.shields.io/pypi/v/star-magic-program.svg?cacheBust=0.395.0)](https://pypi.org/project/star-magic-program/)
-[![Python versions](https://img.shields.io/pypi/pyversions/star-magic-program.svg?cacheBust=0.395.0)](https://pypi.org/project/star-magic-program/)
+[![PyPI version](https://img.shields.io/pypi/v/star-magic-program.svg?cacheBust=0.396.0)](https://pypi.org/project/star-magic-program/)
+[![Python versions](https://img.shields.io/pypi/pyversions/star-magic-program.svg?cacheBust=0.396.0)](https://pypi.org/project/star-magic-program/)
 [![Documentation Status](https://readthedocs.org/projects/star-magic-program/badge/?version=latest)](https://star-magic-program.readthedocs.io/en/latest/?badge=latest)
 [![License: AGPL-3.0 + Commercial](https://img.shields.io/badge/License-AGPL--3.0%20%2B%20Commercial-blue.svg)](LICENSE)
-[![Fidelity gate](https://img.shields.io/badge/fidelity_gate-5801%2F0-brightgreen)](uqff_fidelity_tests.py)
+[![Fidelity gate](https://img.shields.io/badge/fidelity_gate-5816%2F0-brightgreen)](uqff_fidelity_tests.py)
 [![Public surfaces](https://img.shields.io/badge/public_surfaces-2307-blue)](uqff_calculator.py)
 [![Whitepapers](https://img.shields.io/badge/whitepapers-2291-orange)](whitepapers/)
 
-**UQFF systematic rebuild — v0.395.0 complete-compile campaign live**
+**UQFF systematic rebuild — v0.396.0 complete-compile campaign live**
 
-**This release (v0.395.0): THE INSTRUMENT SHIP — THE SIMULATOR BECOMES A COMPLETE INSTRUMENT.** **PAPER_2256 + `uqff_downhole_simulator` v1.6.0:** the program's first packaged industry-application module, built from Daniel's 22Aug2026 template thread (grok_cce7a73b) on same-day GO — a deep-well HPHT simulator (TD ≈ 20,300 ft) with a quartz P/T gauge string whose drift is **suppressed by the canonical lattice**: suppression = (0.58+0.32·(1−F_TRZ))·(0.52+0.38·K_MEX)·(0.68+0.27·Φ_res) = **1.0324 at the locked {0.1, 25/12, 0.84}** — drift BELOW the 0.215 %FS/yr industry baseline. **The knob ruling (Rule 2):** the template's adjustable "K_MEX"/"Φ_res" sliders were tuning gains wearing primitive names — canonicals now LOCKED in-physics (verified live by the dispatch on every call), sliders renamed `k_structural_trim`/`phi_coupling_trim`. **v1.1.0 extensions (Daniel-directed):** N-gauge strings (`make_sensor_string`); real well profiles from CSV (`load_well_profile_csv`; the shipped sample carries an HPHT overpressure kick the linear model cannot see — 18,405 vs 9,313 psi at the deepest gauge); and **twin-gauge drift-comparison mode** — every station carries a conventional reference gauge, and away from the clip band the measured conventional/UQFF ratio EQUALS the canonical suppression (1.0325 measured vs 1.0324 predicted across a 12-gauge string on the real profile) — the PAPER_2250 quartz bench test's complete simulation instrument. **v1.2.0 extension:** service-life drift ACCUMULATION (`ServiceLifeSimulator`) — the rates integrated over simulated service years, twin-leg divergence curves (2-3 psi/yr separation per station at FS 30,000 psi, 10-15 psi at a 5-yr horizon), recalibration resets, years-to-error-budget arithmetic; the accumulated-error ratio converges to the canonical suppression at every station. **v1.3.0 extension:** field-telemetry realism (`TelemetryRecorder`) — 1-reading/min timestamped acquisition with line dropouts, stuck gauges and spikes, historian quality flags, and a scored QC pipeline (frozen-value detection 1.0/1.0; Hampel despiker with common-mode veto, spike precision 0.83-1.0) exporting field-historian-style CSVs. **v1.4.0 extension:** depth-sweep case-study mode (`case_study`/`write_markdown` + CLI) — where the advantage is largest: 2.04 psi/yr flat below the HPHT knees compounding to 3.04 psi/yr at 20,000 ft (4.21 at 27,000 ft on a 30,000-ft well), rendered as a one-page customer case with the honest DERIVED_HYBRID note. **v1.5.0 extension:** real-datasheet gauge specs (`GaugeSpec`/`GAUGE_SPECS`) — web-verified GEO PSI GEOQ 177 (Quartzdyne) presets with mandatory citations (uncited specs rejected in code), plumbed through every layer; the suppression ratio is baseline-independent while the psi/yr story honestly scales with the cited spec bound. **v1.6.0 extensions:** wellbore deviation (MD/TVD surveys — a 60-deg tangent reads 6,525 vs the vertical model's 9,315 psi at the deepest gauge), multi-well batch runs (ratio invariant across geometries), and a headless CLI (`python -m uqff_downhole_simulator run|service-life|telemetry|case-study`). The 2026-08-23 extension list is complete. Thirteen package files (11 modules incl. the headless CLI + README + sample profile), installable package, engine headless by design, DERIVED_HYBRID with every industry anchor inline-commented, gate-verified per run. **Totals: 2,252 wired (2,307 DISPATCH keys) / gate 5,801 green / 4,179 defs. Next paper: PAPER_2257.**
+**This release (v0.396.0): THE TWO-STREAM SHIP — THE ARCHITECTURE COMPLETE.** **PAPER_2256 + `uqff_downhole_simulator` v1.6.0:** the program's first packaged industry-application module, built from Daniel's 22Aug2026 template thread (grok_cce7a73b) on same-day GO — a deep-well HPHT simulator (TD ≈ 20,300 ft) with a quartz P/T gauge string whose drift is **suppressed by the canonical lattice**: suppression = (0.58+0.32·(1−F_TRZ))·(0.52+0.38·K_MEX)·(0.68+0.27·Φ_res) = **1.0324 at the locked {0.1, 25/12, 0.84}** — drift BELOW the 0.215 %FS/yr industry baseline. **The knob ruling (Rule 2):** the template's adjustable "K_MEX"/"Φ_res" sliders were tuning gains wearing primitive names — canonicals now LOCKED in-physics (verified live by the dispatch on every call), sliders renamed `k_structural_trim`/`phi_coupling_trim`. **v1.1.0 extensions (Daniel-directed):** N-gauge strings (`make_sensor_string`); real well profiles from CSV (`load_well_profile_csv`; the shipped sample carries an HPHT overpressure kick the linear model cannot see — 18,405 vs 9,313 psi at the deepest gauge); and **twin-gauge drift-comparison mode** — every station carries a conventional reference gauge, and away from the clip band the measured conventional/UQFF ratio EQUALS the canonical suppression (1.0325 measured vs 1.0324 predicted across a 12-gauge string on the real profile) — the PAPER_2250 quartz bench test's complete simulation instrument. **v1.2.0 extension:** service-life drift ACCUMULATION (`ServiceLifeSimulator`) — the rates integrated over simulated service years, twin-leg divergence curves (2-3 psi/yr separation per station at FS 30,000 psi, 10-15 psi at a 5-yr horizon), recalibration resets, years-to-error-budget arithmetic; the accumulated-error ratio converges to the canonical suppression at every station. **v1.3.0 extension:** field-telemetry realism (`TelemetryRecorder`) — 1-reading/min timestamped acquisition with line dropouts, stuck gauges and spikes, historian quality flags, and a scored QC pipeline (frozen-value detection 1.0/1.0; Hampel despiker with common-mode veto, spike precision 0.83-1.0) exporting field-historian-style CSVs. **v1.4.0 extension:** depth-sweep case-study mode (`case_study`/`write_markdown` + CLI) — where the advantage is largest: 2.04 psi/yr flat below the HPHT knees compounding to 3.04 psi/yr at 20,000 ft (4.21 at 27,000 ft on a 30,000-ft well), rendered as a one-page customer case with the honest DERIVED_HYBRID note. **v1.5.0 extension:** real-datasheet gauge specs (`GaugeSpec`/`GAUGE_SPECS`) — web-verified GEO PSI GEOQ 177 (Quartzdyne) presets with mandatory citations (uncited specs rejected in code), plumbed through every layer; the suppression ratio is baseline-independent while the psi/yr story honestly scales with the cited spec bound. **v1.6.0 extensions:** wellbore deviation (MD/TVD surveys — a 60-deg tangent reads 6,525 vs the vertical model's 9,315 psi at the deepest gauge), multi-well batch runs (ratio invariant across geometries), and a headless CLI (`python -m uqff_downhole_simulator run|service-life|telemetry|case-study`). The 2026-08-23 extension list is complete. **v1.7.0 (two-stream build, piece 1): the tool library** — cited catalog of 8 toolstring entries (quartz twins, piezoresistive class with ChampionX-cited exponential drift form, vibrating wire/DTS/thermocouple as PARAMETERS_USER_SUPPLIED refusing invented numbers, and the G6 Modbus-RS485 surface interface declared as the ports-layer target) + ToolString composition + station rating checks (catches a 177 °C gauge in the 217 °C kick zone). **v1.8.0 (two-stream build, piece 2): the ports/plug-in layer** — READ-ONLY ingest to the normalized LiveStream form: historian-CSV and LAS 2.0 readers IMPLEMENTED (round-trip verified against the v1.3.0 telemetry export; wrapped LAS refused, not mis-parsed), live-protocol taps (modbus_g6/witsml/opcua) DECLARED and refusing until site details exist, `register_port()` for site plug-ins, CLI `ingest` subcommand. **v1.9.0 (piece 3 — the architecture COMPLETE): the two-stream reconciler** — per-station offset series classified IN_FAMILY / CALIBRATION_OFFSET / DRIFT_CONSISTENT / TRANSIENTS / UNEXPLAINED with disclosed thresholds; validated on four scenarios incl. THE FIND: the kick well reconciled against a linear-gradient assumption flags 610/4,448/6,083 psi unexplained offsets — the closed stream exposing the undervalued data stream. CLI `reconcile` subcommand. Sixteen package files (14 modules incl. the headless CLI + README + sample profile), installable package, engine headless by design, DERIVED_HYBRID with every industry anchor inline-commented, gate-verified per run. **Totals: 2,252 wired (2,307 DISPATCH keys) / gate 5,816 green / 4,179 defs. Next paper: PAPER_2257.**
 
 Author: Daniel T. Murphy · Star-Magic Research Program
 License: AGPL-3.0-or-later OR Commercial
@@ -102,7 +102,7 @@ Mizuno, Rossi).
 Full framework physics lives in the whitepaper corpus — the physics is the
 whitepapers; the calculator computes what the whitepapers derive.
 
-## What is currently shipped (v0.395.0)
+## What is currently shipped (v0.396.0)
 
 ### Wiring campaign — LIVE (see `CLAUDE.md` charter)
 

{star_magic_program-0.395.0 → star_magic_program-0.396.0}/README.md

@@ -1,16 +1,16 @@
 # Star-Magic-Program
 
-[![PyPI version](https://img.shields.io/pypi/v/star-magic-program.svg?cacheBust=0.395.0)](https://pypi.org/project/star-magic-program/)
-[![Python versions](https://img.shields.io/pypi/pyversions/star-magic-program.svg?cacheBust=0.395.0)](https://pypi.org/project/star-magic-program/)
+[![PyPI version](https://img.shields.io/pypi/v/star-magic-program.svg?cacheBust=0.396.0)](https://pypi.org/project/star-magic-program/)
+[![Python versions](https://img.shields.io/pypi/pyversions/star-magic-program.svg?cacheBust=0.396.0)](https://pypi.org/project/star-magic-program/)
 [![Documentation Status](https://readthedocs.org/projects/star-magic-program/badge/?version=latest)](https://star-magic-program.readthedocs.io/en/latest/?badge=latest)
 [![License: AGPL-3.0 + Commercial](https://img.shields.io/badge/License-AGPL--3.0%20%2B%20Commercial-blue.svg)](LICENSE)
-[![Fidelity gate](https://img.shields.io/badge/fidelity_gate-5801%2F0-brightgreen)](uqff_fidelity_tests.py)
+[![Fidelity gate](https://img.shields.io/badge/fidelity_gate-5816%2F0-brightgreen)](uqff_fidelity_tests.py)
 [![Public surfaces](https://img.shields.io/badge/public_surfaces-2307-blue)](uqff_calculator.py)
 [![Whitepapers](https://img.shields.io/badge/whitepapers-2291-orange)](whitepapers/)
 
-**UQFF systematic rebuild — v0.395.0 complete-compile campaign live**
+**UQFF systematic rebuild — v0.396.0 complete-compile campaign live**
 
-**This release (v0.395.0): THE INSTRUMENT SHIP — THE SIMULATOR BECOMES A COMPLETE INSTRUMENT.** **PAPER_2256 + `uqff_downhole_simulator` v1.6.0:** the program's first packaged industry-application module, built from Daniel's 22Aug2026 template thread (grok_cce7a73b) on same-day GO — a deep-well HPHT simulator (TD ≈ 20,300 ft) with a quartz P/T gauge string whose drift is **suppressed by the canonical lattice**: suppression = (0.58+0.32·(1−F_TRZ))·(0.52+0.38·K_MEX)·(0.68+0.27·Φ_res) = **1.0324 at the locked {0.1, 25/12, 0.84}** — drift BELOW the 0.215 %FS/yr industry baseline. **The knob ruling (Rule 2):** the template's adjustable "K_MEX"/"Φ_res" sliders were tuning gains wearing primitive names — canonicals now LOCKED in-physics (verified live by the dispatch on every call), sliders renamed `k_structural_trim`/`phi_coupling_trim`. **v1.1.0 extensions (Daniel-directed):** N-gauge strings (`make_sensor_string`); real well profiles from CSV (`load_well_profile_csv`; the shipped sample carries an HPHT overpressure kick the linear model cannot see — 18,405 vs 9,313 psi at the deepest gauge); and **twin-gauge drift-comparison mode** — every station carries a conventional reference gauge, and away from the clip band the measured conventional/UQFF ratio EQUALS the canonical suppression (1.0325 measured vs 1.0324 predicted across a 12-gauge string on the real profile) — the PAPER_2250 quartz bench test's complete simulation instrument. **v1.2.0 extension:** service-life drift ACCUMULATION (`ServiceLifeSimulator`) — the rates integrated over simulated service years, twin-leg divergence curves (2-3 psi/yr separation per station at FS 30,000 psi, 10-15 psi at a 5-yr horizon), recalibration resets, years-to-error-budget arithmetic; the accumulated-error ratio converges to the canonical suppression at every station. **v1.3.0 extension:** field-telemetry realism (`TelemetryRecorder`) — 1-reading/min timestamped acquisition with line dropouts, stuck gauges and spikes, historian quality flags, and a scored QC pipeline (frozen-value detection 1.0/1.0; Hampel despiker with common-mode veto, spike precision 0.83-1.0) exporting field-historian-style CSVs. **v1.4.0 extension:** depth-sweep case-study mode (`case_study`/`write_markdown` + CLI) — where the advantage is largest: 2.04 psi/yr flat below the HPHT knees compounding to 3.04 psi/yr at 20,000 ft (4.21 at 27,000 ft on a 30,000-ft well), rendered as a one-page customer case with the honest DERIVED_HYBRID note. **v1.5.0 extension:** real-datasheet gauge specs (`GaugeSpec`/`GAUGE_SPECS`) — web-verified GEO PSI GEOQ 177 (Quartzdyne) presets with mandatory citations (uncited specs rejected in code), plumbed through every layer; the suppression ratio is baseline-independent while the psi/yr story honestly scales with the cited spec bound. **v1.6.0 extensions:** wellbore deviation (MD/TVD surveys — a 60-deg tangent reads 6,525 vs the vertical model's 9,315 psi at the deepest gauge), multi-well batch runs (ratio invariant across geometries), and a headless CLI (`python -m uqff_downhole_simulator run|service-life|telemetry|case-study`). The 2026-08-23 extension list is complete. Thirteen package files (11 modules incl. the headless CLI + README + sample profile), installable package, engine headless by design, DERIVED_HYBRID with every industry anchor inline-commented, gate-verified per run. **Totals: 2,252 wired (2,307 DISPATCH keys) / gate 5,801 green / 4,179 defs. Next paper: PAPER_2257.**
+**This release (v0.396.0): THE TWO-STREAM SHIP — THE ARCHITECTURE COMPLETE.** **PAPER_2256 + `uqff_downhole_simulator` v1.6.0:** the program's first packaged industry-application module, built from Daniel's 22Aug2026 template thread (grok_cce7a73b) on same-day GO — a deep-well HPHT simulator (TD ≈ 20,300 ft) with a quartz P/T gauge string whose drift is **suppressed by the canonical lattice**: suppression = (0.58+0.32·(1−F_TRZ))·(0.52+0.38·K_MEX)·(0.68+0.27·Φ_res) = **1.0324 at the locked {0.1, 25/12, 0.84}** — drift BELOW the 0.215 %FS/yr industry baseline. **The knob ruling (Rule 2):** the template's adjustable "K_MEX"/"Φ_res" sliders were tuning gains wearing primitive names — canonicals now LOCKED in-physics (verified live by the dispatch on every call), sliders renamed `k_structural_trim`/`phi_coupling_trim`. **v1.1.0 extensions (Daniel-directed):** N-gauge strings (`make_sensor_string`); real well profiles from CSV (`load_well_profile_csv`; the shipped sample carries an HPHT overpressure kick the linear model cannot see — 18,405 vs 9,313 psi at the deepest gauge); and **twin-gauge drift-comparison mode** — every station carries a conventional reference gauge, and away from the clip band the measured conventional/UQFF ratio EQUALS the canonical suppression (1.0325 measured vs 1.0324 predicted across a 12-gauge string on the real profile) — the PAPER_2250 quartz bench test's complete simulation instrument. **v1.2.0 extension:** service-life drift ACCUMULATION (`ServiceLifeSimulator`) — the rates integrated over simulated service years, twin-leg divergence curves (2-3 psi/yr separation per station at FS 30,000 psi, 10-15 psi at a 5-yr horizon), recalibration resets, years-to-error-budget arithmetic; the accumulated-error ratio converges to the canonical suppression at every station. **v1.3.0 extension:** field-telemetry realism (`TelemetryRecorder`) — 1-reading/min timestamped acquisition with line dropouts, stuck gauges and spikes, historian quality flags, and a scored QC pipeline (frozen-value detection 1.0/1.0; Hampel despiker with common-mode veto, spike precision 0.83-1.0) exporting field-historian-style CSVs. **v1.4.0 extension:** depth-sweep case-study mode (`case_study`/`write_markdown` + CLI) — where the advantage is largest: 2.04 psi/yr flat below the HPHT knees compounding to 3.04 psi/yr at 20,000 ft (4.21 at 27,000 ft on a 30,000-ft well), rendered as a one-page customer case with the honest DERIVED_HYBRID note. **v1.5.0 extension:** real-datasheet gauge specs (`GaugeSpec`/`GAUGE_SPECS`) — web-verified GEO PSI GEOQ 177 (Quartzdyne) presets with mandatory citations (uncited specs rejected in code), plumbed through every layer; the suppression ratio is baseline-independent while the psi/yr story honestly scales with the cited spec bound. **v1.6.0 extensions:** wellbore deviation (MD/TVD surveys — a 60-deg tangent reads 6,525 vs the vertical model's 9,315 psi at the deepest gauge), multi-well batch runs (ratio invariant across geometries), and a headless CLI (`python -m uqff_downhole_simulator run|service-life|telemetry|case-study`). The 2026-08-23 extension list is complete. **v1.7.0 (two-stream build, piece 1): the tool library** — cited catalog of 8 toolstring entries (quartz twins, piezoresistive class with ChampionX-cited exponential drift form, vibrating wire/DTS/thermocouple as PARAMETERS_USER_SUPPLIED refusing invented numbers, and the G6 Modbus-RS485 surface interface declared as the ports-layer target) + ToolString composition + station rating checks (catches a 177 °C gauge in the 217 °C kick zone). **v1.8.0 (two-stream build, piece 2): the ports/plug-in layer** — READ-ONLY ingest to the normalized LiveStream form: historian-CSV and LAS 2.0 readers IMPLEMENTED (round-trip verified against the v1.3.0 telemetry export; wrapped LAS refused, not mis-parsed), live-protocol taps (modbus_g6/witsml/opcua) DECLARED and refusing until site details exist, `register_port()` for site plug-ins, CLI `ingest` subcommand. **v1.9.0 (piece 3 — the architecture COMPLETE): the two-stream reconciler** — per-station offset series classified IN_FAMILY / CALIBRATION_OFFSET / DRIFT_CONSISTENT / TRANSIENTS / UNEXPLAINED with disclosed thresholds; validated on four scenarios incl. THE FIND: the kick well reconciled against a linear-gradient assumption flags 610/4,448/6,083 psi unexplained offsets — the closed stream exposing the undervalued data stream. CLI `reconcile` subcommand. Sixteen package files (14 modules incl. the headless CLI + README + sample profile), installable package, engine headless by design, DERIVED_HYBRID with every industry anchor inline-commented, gate-verified per run. **Totals: 2,252 wired (2,307 DISPATCH keys) / gate 5,816 green / 4,179 defs. Next paper: PAPER_2257.**
 
 Author: Daniel T. Murphy · Star-Magic Research Program
 License: AGPL-3.0-or-later OR Commercial
@@ -68,7 +68,7 @@ Mizuno, Rossi).
 Full framework physics lives in the whitepaper corpus — the physics is the
 whitepapers; the calculator computes what the whitepapers derive.
 
-## What is currently shipped (v0.395.0)
+## What is currently shipped (v0.396.0)
 
 ### Wiring campaign — LIVE (see `CLAUDE.md` charter)
 

{star_magic_program-0.395.0 → star_magic_program-0.396.0}/SESSION_LOG.md

@@ -11811,3 +11811,40 @@ own source for quote-adjacent Unix temp-path literals (needle constructed at
 runtime so the guard cannot self-match) — the mistake cannot be reintroduced.
 Gate 5,800 → 5,801; README badge + pyproject description synced. Standing
 lesson for the charter set: gate pins write scratch files ONLY via tempfile.
+
+## Entry 268 — 2026-08-24 — v0.396.0 THE TWO-STREAM SHIP (downhole v1.7.0–v1.9.0)
+
+Daniel's architectural ruling (2026-08-23, verbatim frame): "There are two
+sides to this program so that the standalone theoretical can function to find
+the offset undervalued data streams that coordinate live stream with closed
+stream." Confirmed correct reading; built as three pieces, COMPLETE:
+
+1. v1.7.0 `uqff_tool_library.py` — cited 8-entry catalog: quartz twins (GEOQ
+   177 specs), piezoresistive class (ChampionX-cited exponential-in-T FORM,
+   representative coefficients disclosed), GEOVW/GEOXTR/GEOPulse as
+   PARAMETERS_USER_SUPPLIED with drift_model_for() REFUSING them, G6 Modbus
+   interface = declared ports target; ToolString + rating_check (177 °C gauge
+   caught at the 217 °C kick station).
+2. v1.8.0 `uqff_ports.py` — READ-ONLY plug-in registry → normalized
+   LiveStream. historian_csv + las2 IMPLEMENTED; round-trip vs the v1.3.0
+   telemetry export verified to precision (MISSING→NaN, flags carried);
+   wrapped LAS REFUSED; modbus_g6/witsml/opcua DECLARED-refusing;
+   register_port(); CLI ingest.
+3. v1.9.0 `uqff_reconciler.py` — offset(t) = measured − predicted per
+   station; classifications with thresholds DISCLOSED in every report; drift
+   not classified under ~18-day windows (slopes are noise, said so). Four
+   scenarios verified per gate run: clean 6/6 IN_FAMILY; +50 psi bias →
+   CALIBRATION_OFFSET 50.09 recovered; 2-yr synthetic at conventional rate →
+   DRIFT_CONSISTENT (82.45 in [79.71, 82.29]); THE FIND — kick well vs
+   linear-gradient assumption → UNEXPLAINED_OFFSET 610/4,448/6,083 psi.
+
+Session also carried the v0.395.0 RED-GATE portability incident (hard-coded
+Unix temp path in a gate pin; fixed; permanent PORTABILITY GUARD self-scan
+added — gate 5,800→5,801) before this arc (5,801→5,816, +15 pins).
+
+Ship prep: 6 pins at 0.396.0 (desc 495 chars, measured counts); SHIP GUARD
+marker INSTRUMENT_ARC→TWOSTREAM_ARC (+ GAPS id twostream_arc_rule7); README
+release paragraph/campaign/shipped header; CHANGELOG + _BUILD_LOG +
+SHIP_MESSAGE + RULINGS trail; 10 TWOSTREAM_ARC trail rows. Zero calculator
+physics changes; zero new dispatches (ratchet untouched). Awaiting Daniel:
+`.\ship.ps1` (clear .git/index.lock first).

star_magic_program-0.396.0/UNIFIED_REGISTRY_VERSION.txt

@@ -0,0 +1 @@
+v0.396.0

{star_magic_program-0.395.0 → star_magic_program-0.396.0}/UNIFIED_REGISTRY_XGEO_QUEUE.csv

@@ -1722,3 +1722,4 @@ CENSUS_ARC,n/a,n/a,NOT_APPLICABLE,census consolidation single-geometry,v0.392.0
 LEDGER_ARC,n/a,n/a,NOT_APPLICABLE,census+packaging single-geometry,v0.393.0
 DOWNHOLE_ARC,n/a,n/a,NOT_APPLICABLE,application module single-geometry,v0.394.0
 INSTRUMENT_ARC,n/a,n/a,NOT_APPLICABLE,application-module extensions single-geometry,v0.395.0
+TWOSTREAM_ARC,n/a,n/a,NOT_APPLICABLE,application-module architecture single-geometry,v0.396.0

{star_magic_program-0.395.0 → star_magic_program-0.396.0}/UNIFIED_REGISTRY_XGEO_ROUTES.csv

@@ -1723,3 +1723,4 @@ CENSUS_ARC,n/a,XGEO_NOT_REQUIRED,none,v0.392.0
 LEDGER_ARC,n/a,XGEO_NOT_REQUIRED,none,v0.393.0
 DOWNHOLE_ARC,n/a,XGEO_NOT_REQUIRED,none,v0.394.0
 INSTRUMENT_ARC,n/a,XGEO_NOT_REQUIRED,none,v0.395.0
+TWOSTREAM_ARC,n/a,XGEO_NOT_REQUIRED,none,v0.396.0

{star_magic_program-0.395.0 → star_magic_program-0.396.0}/_BUILD_LOG.md

@@ -306,3 +306,6 @@ THE DOWNHOLE SHIP: PAPER_2256 + uqff_downhole_simulator v1.1.0 (first industry-a
 
 ## v0.395.0 — 2026-08-23
 THE INSTRUMENT SHIP: uqff_downhole_simulator v1.2.0-v1.6.0 (service-life divergence curves; telemetry realism with scored QC; depth-sweep case studies; web-verified GEOQ 177 gauge specs with mandatory citations; MD/TVD deviation; batch runs; headless CLI). Suppression ratio 1.0324 invariant across geometry and baseline. Gate 5,800/0. 23-file pass verified.
+
+## v0.396.0 — 2026-08-24
+THE TWO-STREAM SHIP: uqff_downhole_simulator v1.7.0-v1.9.0 (cited tool library; read-only ports with historian/LAS implemented and live taps declared; two-stream reconciler with four-scenario validation incl. the 6,083-psi kick find). Architecture complete. Gate 5,816/0. 23-file pass verified.

{star_magic_program-0.395.0 → star_magic_program-0.396.0}/pyproject.toml

@@ -4,8 +4,8 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "star-magic-program"
-version = "0.395.0"
-description = "Star-Magic UQFF calculator v0.395.0: THE INSTRUMENT SHIP. 2,252 dispatches (2,307 keys), gate 5,801/0. uqff_downhole_simulator v1.2.0-v1.6.0: service-life divergence curves, field-telemetry realism with scored QC, depth-sweep customer case studies, web-verified datasheet gauge specs (GEO PSI GEOQ 177 / Quartzdyne), MD/TVD wellbore deviation, multi-well batch runs, headless CLI. Suppression ratio 1.0324 invariant across geometry and baseline. Next: PAPER_2257."
+version = "0.396.0"
+description = "Star-Magic UQFF calculator v0.396.0: THE TWO-STREAM SHIP. 2,252 dispatches (2,307 keys), gate 5,816/0. uqff_downhole_simulator v1.7.0-v1.9.0 completes Daniel's two-stream architecture: cited tool library (8 entries, no invented vendor data), read-only ports (historian CSV + LAS 2.0 implemented, live taps declared), and the two-stream reconciler - closed-stream predictions vs live data, offsets classified, undervalued streams flagged (kick-well find: 6,083 psi unexplained). Next: PAPER_2257."
 readme = "README.md"
 authors = [
     { name = "Daniel T. Murphy", email = "daniel.murphy00@enrgyone.com" }

{star_magic_program-0.395.0 → star_magic_program-0.396.0/star_magic_program.egg-info}/PKG-INFO

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: star-magic-program
-Version: 0.395.0
-Summary: Star-Magic UQFF calculator v0.395.0: THE INSTRUMENT SHIP. 2,252 dispatches (2,307 keys), gate 5,801/0. uqff_downhole_simulator v1.2.0-v1.6.0: service-life divergence curves, field-telemetry realism with scored QC, depth-sweep customer case studies, web-verified datasheet gauge specs (GEO PSI GEOQ 177 / Quartzdyne), MD/TVD wellbore deviation, multi-well batch runs, headless CLI. Suppression ratio 1.0324 invariant across geometry and baseline. Next: PAPER_2257.
+Version: 0.396.0
+Summary: Star-Magic UQFF calculator v0.396.0: THE TWO-STREAM SHIP. 2,252 dispatches (2,307 keys), gate 5,816/0. uqff_downhole_simulator v1.7.0-v1.9.0 completes Daniel's two-stream architecture: cited tool library (8 entries, no invented vendor data), read-only ports (historian CSV + LAS 2.0 implemented, live taps declared), and the two-stream reconciler - closed-stream predictions vs live data, offsets classified, undervalued streams flagged (kick-well find: 6,083 psi unexplained). Next: PAPER_2257.
 Author-email: "Daniel T. Murphy" <daniel.murphy00@enrgyone.com>
 License: AGPL-3.0-or-later OR LicenseRef-StarMagic-Commercial
 Project-URL: Homepage, https://github.com/Daniel8Murphy0007/Star-Magic-Program
@@ -34,17 +34,17 @@ Dynamic: license-file
 
 # Star-Magic-Program
 
-[![PyPI version](https://img.shields.io/pypi/v/star-magic-program.svg?cacheBust=0.395.0)](https://pypi.org/project/star-magic-program/)
-[![Python versions](https://img.shields.io/pypi/pyversions/star-magic-program.svg?cacheBust=0.395.0)](https://pypi.org/project/star-magic-program/)
+[![PyPI version](https://img.shields.io/pypi/v/star-magic-program.svg?cacheBust=0.396.0)](https://pypi.org/project/star-magic-program/)
+[![Python versions](https://img.shields.io/pypi/pyversions/star-magic-program.svg?cacheBust=0.396.0)](https://pypi.org/project/star-magic-program/)
 [![Documentation Status](https://readthedocs.org/projects/star-magic-program/badge/?version=latest)](https://star-magic-program.readthedocs.io/en/latest/?badge=latest)
 [![License: AGPL-3.0 + Commercial](https://img.shields.io/badge/License-AGPL--3.0%20%2B%20Commercial-blue.svg)](LICENSE)
-[![Fidelity gate](https://img.shields.io/badge/fidelity_gate-5801%2F0-brightgreen)](uqff_fidelity_tests.py)
+[![Fidelity gate](https://img.shields.io/badge/fidelity_gate-5816%2F0-brightgreen)](uqff_fidelity_tests.py)
 [![Public surfaces](https://img.shields.io/badge/public_surfaces-2307-blue)](uqff_calculator.py)
 [![Whitepapers](https://img.shields.io/badge/whitepapers-2291-orange)](whitepapers/)
 
-**UQFF systematic rebuild — v0.395.0 complete-compile campaign live**
+**UQFF systematic rebuild — v0.396.0 complete-compile campaign live**
 
-**This release (v0.395.0): THE INSTRUMENT SHIP — THE SIMULATOR BECOMES A COMPLETE INSTRUMENT.** **PAPER_2256 + `uqff_downhole_simulator` v1.6.0:** the program's first packaged industry-application module, built from Daniel's 22Aug2026 template thread (grok_cce7a73b) on same-day GO — a deep-well HPHT simulator (TD ≈ 20,300 ft) with a quartz P/T gauge string whose drift is **suppressed by the canonical lattice**: suppression = (0.58+0.32·(1−F_TRZ))·(0.52+0.38·K_MEX)·(0.68+0.27·Φ_res) = **1.0324 at the locked {0.1, 25/12, 0.84}** — drift BELOW the 0.215 %FS/yr industry baseline. **The knob ruling (Rule 2):** the template's adjustable "K_MEX"/"Φ_res" sliders were tuning gains wearing primitive names — canonicals now LOCKED in-physics (verified live by the dispatch on every call), sliders renamed `k_structural_trim`/`phi_coupling_trim`. **v1.1.0 extensions (Daniel-directed):** N-gauge strings (`make_sensor_string`); real well profiles from CSV (`load_well_profile_csv`; the shipped sample carries an HPHT overpressure kick the linear model cannot see — 18,405 vs 9,313 psi at the deepest gauge); and **twin-gauge drift-comparison mode** — every station carries a conventional reference gauge, and away from the clip band the measured conventional/UQFF ratio EQUALS the canonical suppression (1.0325 measured vs 1.0324 predicted across a 12-gauge string on the real profile) — the PAPER_2250 quartz bench test's complete simulation instrument. **v1.2.0 extension:** service-life drift ACCUMULATION (`ServiceLifeSimulator`) — the rates integrated over simulated service years, twin-leg divergence curves (2-3 psi/yr separation per station at FS 30,000 psi, 10-15 psi at a 5-yr horizon), recalibration resets, years-to-error-budget arithmetic; the accumulated-error ratio converges to the canonical suppression at every station. **v1.3.0 extension:** field-telemetry realism (`TelemetryRecorder`) — 1-reading/min timestamped acquisition with line dropouts, stuck gauges and spikes, historian quality flags, and a scored QC pipeline (frozen-value detection 1.0/1.0; Hampel despiker with common-mode veto, spike precision 0.83-1.0) exporting field-historian-style CSVs. **v1.4.0 extension:** depth-sweep case-study mode (`case_study`/`write_markdown` + CLI) — where the advantage is largest: 2.04 psi/yr flat below the HPHT knees compounding to 3.04 psi/yr at 20,000 ft (4.21 at 27,000 ft on a 30,000-ft well), rendered as a one-page customer case with the honest DERIVED_HYBRID note. **v1.5.0 extension:** real-datasheet gauge specs (`GaugeSpec`/`GAUGE_SPECS`) — web-verified GEO PSI GEOQ 177 (Quartzdyne) presets with mandatory citations (uncited specs rejected in code), plumbed through every layer; the suppression ratio is baseline-independent while the psi/yr story honestly scales with the cited spec bound. **v1.6.0 extensions:** wellbore deviation (MD/TVD surveys — a 60-deg tangent reads 6,525 vs the vertical model's 9,315 psi at the deepest gauge), multi-well batch runs (ratio invariant across geometries), and a headless CLI (`python -m uqff_downhole_simulator run|service-life|telemetry|case-study`). The 2026-08-23 extension list is complete. Thirteen package files (11 modules incl. the headless CLI + README + sample profile), installable package, engine headless by design, DERIVED_HYBRID with every industry anchor inline-commented, gate-verified per run. **Totals: 2,252 wired (2,307 DISPATCH keys) / gate 5,801 green / 4,179 defs. Next paper: PAPER_2257.**
+**This release (v0.396.0): THE TWO-STREAM SHIP — THE ARCHITECTURE COMPLETE.** **PAPER_2256 + `uqff_downhole_simulator` v1.6.0:** the program's first packaged industry-application module, built from Daniel's 22Aug2026 template thread (grok_cce7a73b) on same-day GO — a deep-well HPHT simulator (TD ≈ 20,300 ft) with a quartz P/T gauge string whose drift is **suppressed by the canonical lattice**: suppression = (0.58+0.32·(1−F_TRZ))·(0.52+0.38·K_MEX)·(0.68+0.27·Φ_res) = **1.0324 at the locked {0.1, 25/12, 0.84}** — drift BELOW the 0.215 %FS/yr industry baseline. **The knob ruling (Rule 2):** the template's adjustable "K_MEX"/"Φ_res" sliders were tuning gains wearing primitive names — canonicals now LOCKED in-physics (verified live by the dispatch on every call), sliders renamed `k_structural_trim`/`phi_coupling_trim`. **v1.1.0 extensions (Daniel-directed):** N-gauge strings (`make_sensor_string`); real well profiles from CSV (`load_well_profile_csv`; the shipped sample carries an HPHT overpressure kick the linear model cannot see — 18,405 vs 9,313 psi at the deepest gauge); and **twin-gauge drift-comparison mode** — every station carries a conventional reference gauge, and away from the clip band the measured conventional/UQFF ratio EQUALS the canonical suppression (1.0325 measured vs 1.0324 predicted across a 12-gauge string on the real profile) — the PAPER_2250 quartz bench test's complete simulation instrument. **v1.2.0 extension:** service-life drift ACCUMULATION (`ServiceLifeSimulator`) — the rates integrated over simulated service years, twin-leg divergence curves (2-3 psi/yr separation per station at FS 30,000 psi, 10-15 psi at a 5-yr horizon), recalibration resets, years-to-error-budget arithmetic; the accumulated-error ratio converges to the canonical suppression at every station. **v1.3.0 extension:** field-telemetry realism (`TelemetryRecorder`) — 1-reading/min timestamped acquisition with line dropouts, stuck gauges and spikes, historian quality flags, and a scored QC pipeline (frozen-value detection 1.0/1.0; Hampel despiker with common-mode veto, spike precision 0.83-1.0) exporting field-historian-style CSVs. **v1.4.0 extension:** depth-sweep case-study mode (`case_study`/`write_markdown` + CLI) — where the advantage is largest: 2.04 psi/yr flat below the HPHT knees compounding to 3.04 psi/yr at 20,000 ft (4.21 at 27,000 ft on a 30,000-ft well), rendered as a one-page customer case with the honest DERIVED_HYBRID note. **v1.5.0 extension:** real-datasheet gauge specs (`GaugeSpec`/`GAUGE_SPECS`) — web-verified GEO PSI GEOQ 177 (Quartzdyne) presets with mandatory citations (uncited specs rejected in code), plumbed through every layer; the suppression ratio is baseline-independent while the psi/yr story honestly scales with the cited spec bound. **v1.6.0 extensions:** wellbore deviation (MD/TVD surveys — a 60-deg tangent reads 6,525 vs the vertical model's 9,315 psi at the deepest gauge), multi-well batch runs (ratio invariant across geometries), and a headless CLI (`python -m uqff_downhole_simulator run|service-life|telemetry|case-study`). The 2026-08-23 extension list is complete. **v1.7.0 (two-stream build, piece 1): the tool library** — cited catalog of 8 toolstring entries (quartz twins, piezoresistive class with ChampionX-cited exponential drift form, vibrating wire/DTS/thermocouple as PARAMETERS_USER_SUPPLIED refusing invented numbers, and the G6 Modbus-RS485 surface interface declared as the ports-layer target) + ToolString composition + station rating checks (catches a 177 °C gauge in the 217 °C kick zone). **v1.8.0 (two-stream build, piece 2): the ports/plug-in layer** — READ-ONLY ingest to the normalized LiveStream form: historian-CSV and LAS 2.0 readers IMPLEMENTED (round-trip verified against the v1.3.0 telemetry export; wrapped LAS refused, not mis-parsed), live-protocol taps (modbus_g6/witsml/opcua) DECLARED and refusing until site details exist, `register_port()` for site plug-ins, CLI `ingest` subcommand. **v1.9.0 (piece 3 — the architecture COMPLETE): the two-stream reconciler** — per-station offset series classified IN_FAMILY / CALIBRATION_OFFSET / DRIFT_CONSISTENT / TRANSIENTS / UNEXPLAINED with disclosed thresholds; validated on four scenarios incl. THE FIND: the kick well reconciled against a linear-gradient assumption flags 610/4,448/6,083 psi unexplained offsets — the closed stream exposing the undervalued data stream. CLI `reconcile` subcommand. Sixteen package files (14 modules incl. the headless CLI + README + sample profile), installable package, engine headless by design, DERIVED_HYBRID with every industry anchor inline-commented, gate-verified per run. **Totals: 2,252 wired (2,307 DISPATCH keys) / gate 5,816 green / 4,179 defs. Next paper: PAPER_2257.**
 
 Author: Daniel T. Murphy · Star-Magic Research Program
 License: AGPL-3.0-or-later OR Commercial
@@ -102,7 +102,7 @@ Mizuno, Rossi).
 Full framework physics lives in the whitepaper corpus — the physics is the
 whitepapers; the calculator computes what the whitepapers derive.
 
-## What is currently shipped (v0.395.0)
+## What is currently shipped (v0.396.0)
 
 ### Wiring campaign — LIVE (see `CLAUDE.md` charter)
 

{star_magic_program-0.395.0 → star_magic_program-0.396.0}/star_magic_program.egg-info/SOURCES.txt

@@ -62,6 +62,9 @@ uqff_downhole_simulator/uqff_case_study.py
 uqff_downhole_simulator/uqff_deviation.py
 uqff_downhole_simulator/uqff_downhole_engine.py
 uqff_downhole_simulator/uqff_gauge_specs.py
+uqff_downhole_simulator/uqff_ports.py
 uqff_downhole_simulator/uqff_quartz_hpht_extension.py
+uqff_downhole_simulator/uqff_reconciler.py
 uqff_downhole_simulator/uqff_service_life.py
-uqff_downhole_simulator/uqff_telemetry.py
+uqff_downhole_simulator/uqff_telemetry.py
+uqff_downhole_simulator/uqff_tool_library.py

{star_magic_program-0.395.0 → star_magic_program-0.396.0}/uqff_calculator.py

@@ -73,7 +73,7 @@ from uqff_registry_primitives import (
     DELTA_M2_21_EV2, DELTA_M2_32_EV2,
 )
 
-VERSION = "0.395.0"
+VERSION = "0.396.0"
 
 # NAMED OBSERVED SI ANCHORS (constant drain 2026-08-16, PAPER_2141 bulk pattern + PAPER_2149 observation-headlining)
 # Bit-identical to the literals they replace; UQFF-derived counterparts live in uqff_registry_primitives.

{star_magic_program-0.395.0 → star_magic_program-0.396.0}/uqff_downhole_simulator/README.md

@@ -215,3 +215,79 @@ cs = case_study(CaseStudyConfig(gauge_spec=GAUGE_SPECS['geoq177_30k']))
 python -m uqff_downhole_simulator run --steps 200 --gauges 8 --out run.csv
 python -m uqff_downhole_simulator case-study --td 22000 --kickoff 9000 --inclination 45 --spec geoq177_30k --out case.md
 ```
+
+## v1.7.0 extension: the tool library (Daniel-directed, 2026-08-23 — first piece of the two-stream build)
+
+`uqff_tool_library.py` generalizes the gauge-spec discipline to the whole
+toolstring: a cited catalog (`TOOL_LIBRARY`, 8 entries) of downhole and
+surface tools — UQFF/conventional quartz P/T (wrapping the v1.5.0 GaugeSpecs),
+a piezoresistive class (drift FORM cited from ChampionX: unpredictable,
+exponential in temperature; coefficients disclosed as representative fit),
+vibrating-wire, 18-point thermocouple string, DTS fiber, and the **G6 surface
+interface (Modbus RS485 + 4-20mA)** — every entry declaring the telemetry
+interface it speaks, which is the declaration the live-stream ports/plug-in
+layer will implement.
+
+Honesty structure: entries whose existence is verified but whose numbers were
+not published are `PARAMETERS_USER_SUPPLIED` — their fields are None and
+`drift_model_for()` refuses to invent vendor data. `ToolString` composes
+mixed strings of stations; `rating_check()` checks every tool against the
+well conditions at its station (profile or gradients, deviation honored) —
+it catches a 177 °C gauge hung in the sample well's 217 °C kick zone before
+the well does.
+
+```python
+from uqff_downhole_simulator import ToolString, rating_check, load_well_profile_csv
+ts = ToolString([(3000, 'quartz_pt_uqff_geoq177_30k'), (18200, 'piezoresistive_pt_class')])
+print(rating_check(ts, profile=load_well_profile_csv("uqff_downhole_simulator/sample_well_profile.csv")))
+```
+
+## v1.8.0 extension: the ports/plug-in layer (Daniel-directed, 2026-08-23 — two-stream build, piece 2)
+
+`uqff_ports.py` — READ-ONLY taps ingesting live-stream data into the
+normalized `LiveStream` form the reconciler will consume (index = time or
+depth; per-channel values with NaN for missing; quality flags carried; units
+carried). `PORT_REGISTRY` with explicit statuses: **IMPLEMENTED** —
+`historian_csv` (wide-format historian exports, auto-detecting the v1.3.0
+telemetry layout) and `las2` (LAS 2.0 well logs, CWLS public standard, NULL
+substitution, wrapped mode refused rather than mis-parsed); **DECLARED** —
+`modbus_g6` (the tool library's declared port target), `witsml`, `opcua` —
+these name the protocol but REFUSE to run until real site details exist (no
+invented site behavior). `register_port()` lets a site plug in its own reader
+additively.
+
+The proof loop: the simulator's own v1.3.0 field-style export re-ingested
+through the port matches the recorder's arrays to export precision, with
+MISSING → NaN and flags carried — closed stream → simulated live file →
+port → same numbers. The ingest path is verified in simulation before it
+ever touches a site. CLI: `python -m uqff_downhole_simulator ingest --file
+field.csv` (or `--port las2`).
+
+## v1.9.0 extension: the two-stream reconciler (Daniel-directed, 2026-08-23 — piece 3, the architecture complete)
+
+`uqff_reconciler.py` — the coordinator. The closed stream predicts what every
+gauge in a described well *should* read; a `LiveStream` delivers what it
+*does* read; the reconciler works the per-station offset series and
+classifies each: IN_FAMILY, CALIBRATION_OFFSET (bias recovered),
+DRIFT_CONSISTENT (trend inside the closed stream's own drift envelope
+[UQFF rate, conventional rate] at station T/P — needs ≥18 days of data,
+below that slopes are noise and it says so), TRANSIENTS, or
+**UNEXPLAINED_OFFSET/TREND — the undervalued streams**. Thresholds are
+disclosed engineering heuristics; labels are advisory triage and the numbers
+(bias, slope, sigma, envelope) always ride along.
+
+Validated on four scenarios: clean well → 6/6 IN_FAMILY; +50 psi biased
+gauge → CALIBRATION_OFFSET, 50.09 psi recovered; 2-year synthetic drifting
+at the conventional rate → DRIFT_CONSISTENT (slope 82.45 vs envelope
+[79.71, 82.29] psi/yr); and **the find** — the kick well reconciled against
+a linear-gradient assumption flags 610 / 4,448 / 6,083 psi unexplained
+offsets at the deep stations. The closed stream exposes what the assumed
+model cannot see.
+
+```python
+from uqff_downhole_simulator import Reconciler, SimulatorConfig, ingest
+report = Reconciler(SimulatorConfig(td_ft=18500)).reconcile(ingest("field.csv"))
+print(report['undervalued_streams'])
+```
+
+CLI: `python -m uqff_downhole_simulator reconcile --file field.csv --td 18500`

{star_magic_program-0.395.0 → star_magic_program-0.396.0}/uqff_downhole_simulator/__init__.py

@@ -62,8 +62,31 @@ from .uqff_deviation import (
     load_deviation_csv,
 )
 from .uqff_downhole_engine import run_batch
+from .uqff_tool_library import (
+    ToolSpec,
+    TOOL_LIBRARY,
+    ToolString,
+    drift_model_for,
+    piezoresistive_drift,
+    rating_check,
+)
+from .uqff_ports import (
+    LiveStream,
+    StreamChannel,
+    PortSpec,
+    PORT_REGISTRY,
+    ingest,
+    read_historian_csv,
+    read_las,
+    register_port,
+)
+from .uqff_reconciler import (
+    Reconciler,
+    ReconcilerConfig,
+    auto_station_map,
+)
 
-__version__ = "1.6.0"
+__version__ = "1.9.0"
 __all__ = [
     "calculate_quartz_transducer_hpht_UQFF", "canonical_suppression",
     "conventional_drift", "drift_comparison",
@@ -74,5 +97,10 @@ __all__ = [
     "CaseStudyConfig", "case_study", "depth_sweep", "write_markdown",
     "GaugeSpec", "GAUGE_SPECS", "load_gauge_spec_json",
     "DeviationSurvey", "load_deviation_csv", "run_batch",
+    "ToolSpec", "TOOL_LIBRARY", "ToolString", "drift_model_for",
+    "piezoresistive_drift", "rating_check",
+    "LiveStream", "StreamChannel", "PortSpec", "PORT_REGISTRY",
+    "ingest", "read_historian_csv", "read_las", "register_port",
+    "Reconciler", "ReconcilerConfig", "auto_station_map",
     "DEFAULT_TD_FT", "DEFAULT_SENSOR_DEPTHS_FT",
 ]

{star_magic_program-0.395.0 → star_magic_program-0.396.0}/uqff_downhole_simulator/__main__.py

@@ -70,6 +70,16 @@ def main(argv=None) -> int:
     p_tm.add_argument("--seed", type=int, default=None)
     p_tm.add_argument("--out", type=str, default="uqff_telemetry.csv")
 
+    p_in = sub.add_parser("ingest", help="ingest a live-stream file through a port (read-only)")
+    p_in.add_argument("--file", type=str, required=True, help="source file (historian CSV or LAS)")
+    p_in.add_argument("--port", type=str, default="historian_csv",
+                      help="port name from PORT_REGISTRY (historian_csv | las2 | site plug-ins)")
+
+    p_rc = sub.add_parser("reconcile", help="two-stream reconciliation: live file vs closed-stream prediction")
+    _add_well_args(p_rc)
+    p_rc.add_argument("--file", type=str, required=True, help="live-stream file (historian CSV)")
+    p_rc.add_argument("--port", type=str, default="historian_csv")
+
     p_cs = sub.add_parser("case-study", help="depth sweep, write the one-page markdown case")
     _add_well_args(p_cs)
     p_cs.add_argument("--points", type=int, default=12)
@@ -107,6 +117,16 @@ def main(argv=None) -> int:
         s = rec.telemetry_summary()
         print(f"telemetry: {s['samples']} samples -> {out} "
               f"(uptime {s['uptime_pct']}%, stuck P/R {s['stuck_score']['precision']}/{s['stuck_score']['recall']})")
+    elif a.cmd == "ingest":
+        from . import ingest as _ingest
+        st = _ingest(a.file, port=a.port)
+        import json as _json
+        print(_json.dumps(st.summary(), indent=1))
+    elif a.cmd == "reconcile":
+        from . import ingest as _ingest, Reconciler
+        rep = Reconciler(_build_config(a)).reconcile(_ingest(a.file, port=a.port))
+        import json as _json
+        print(_json.dumps(rep, indent=1))
     elif a.cmd == "case-study":
         from .uqff_deviation import DeviationSurvey
         td = a.td if a.td is not None else DEFAULT_TD_FT