Exact release identity: MathGov_Core_2026_09_v12.6_SGP_v8.5+2026.08.15.3
License: Apache-2.0 except where a file states otherwise
Reality -> Rights -> Ripples. Qualify -> Rank.
MathGov is an open-source, rights-constrained governance framework. Its fixed five-stage cascade is:
RG -> RF/NCRC -> TRC -> CSV -> RLS
v12.6 closes the gap between selecting a qualified option and carrying out a real action. A consequence-bearing action must now remain bound to the option, configuration, transition, post-state, capability state, and qualification snapshot that were actually reviewed. A mismatch blocks execution. Material outcomes or context changes reopen qualification.
The release also strengthens physical-causal warrant boundaries, claim-domain bridges, carried obligations, hidden human compensation checks, source-coupling integrity, and transparent workbook assurance. It adds no sixth gate, no eighth welfare dimension, and no new authority layer.
For consequential computational systems, the Agent System and ripple.md now expose a compact derived view of the computational problem/object, exact architecture/configuration, demonstrated capability, assigned task, deployment domain, evidence standard, consequence interface, and separate legitimate authority basis. The view does not add a gate or create authority. Its critical rule is: Consequence interface is not legitimate authority.
Public stabilization window: 15 August 2026 through 15 February 2027. During this period, Canon v12.6 is frozen against speculative feature accretion. Same-line corrections are limited to demonstrated correctness, rights, interpretability, security, citation/release-integrity defects, or material legal/scientific evidence. New constructs and discretionary extensions belong in the roadmap until supported by a clearly superior formally justified design. The primary development emphasis now shifts to independent replay, inter-rater studies, RLS/WDBIP/UCI measurement work, and bounded institutional pilots.
- Public introduction:
docs/guides/MATHGOV_3R_1_2_PUBLIC_INTRO_v12_6.md - Governing Canon:
docs/canon/RippleLogic_v12.6_Canon.md - SGP:
docs/sgp/SGP_v8.5.md - Exact Core 15 bundle:
core_15/ - Component map:
docs/CORE_COMPONENT_MAP.md - Current adversarial-audit disposition:
release/MATHGOV_V12_6_CLAUDE_ADVERSARIAL_AUDIT_DISPOSITION_2026_08_11.md - Prior change and independence report:
release/MATHGOV_V12_6_AND_MHIOS_V0_7_INDEPENDENT_RELEASE_REPORT.md - Final verification:
release/FINAL_VERIFICATION_REPORT.md - Source verification log:
release/SOURCE_VERIFICATION_LOG.md - Examples and reference replays:
docs/examples/README.md - Negative-example catalog:
docs/examples/NEGATIVE_EXAMPLES_INDEX.md - Evidence-stage maturity ladder:
docs/assurance/SCIENTIFIC_MATURITY_LADDER.md - High-stakes physical/causal worked example:
docs/examples/PC_AEP_HIGH_STAKES_WORKED_EXAMPLE.md - Physical execution boundary guide:
docs/guides/PHYSICAL_ADMISSIBILITY_AND_EXECUTION_BOUNDARY.md - Safety-conformance vectors:
docs/validation/V12_6_SAFETY_CONFORMANCE_VECTORS.md
| Component | Version | Governing source |
|---|---|---|
| RippleLogic Canon | 12.6 | docs/canon/RippleLogic_v12.6_Canon.md |
| Sentience Gradient Protocol | 8.5 | docs/sgp/SGP_v8.5.md |
| ripple.md Standard | 5.5 | docs/standards/ripple_md_Standard_v5.5.md |
| RippleLogic Agent System | 12.5 | docs/agents/RippleLogic_Agent_System_v12.5.md |
| CSV Gate Standard | 2.4 | docs/standards/CSV_Gate_Standard_v2.4.md |
| RippleLogic Cascade Standard | 2.6 | docs/standards/RippleLogic_Cascade_Standard_v2.6.md |
| Reproducibility and Use Standard | 1.4 | docs/implementation/MATHGOV_REPRODUCIBILITY_AND_USE_STANDARD_v1.4.md |
| WDBIP | 1.6 | docs/standards/wdbip/Welfare_Dimension_Boundary_and_Interaction_Protocol_v1.6.md |
| RLS Validation Protocol | 2.6 | docs/validation/rls/RippleLogic_RLS_Validation_Protocol_v2_6.md |
| Foundations Primer | 4.4 | docs/primer/RippleLogic_Foundations_Primer_v4.4.md |
| Public Introduction | 12.6 | docs/guides/MATHGOV_3R_1_2_PUBLIC_INTRO_v12_6.md |
| PC-AEP | 2.3 | docs/standards/Physical_Causal_Admissibility_Evidence_Profile_v2.3.md |
| MFDI Standard | 2.3 | docs/standards/Methodological_Falsifiability_and_Dependency_Integrity_Standard_v2.3.md |
| Source-Coupling Integrity Standard | 2.3 | docs/standards/Source_Coupling_Integrity_Standard_v2.3.md |
| RippleLogic Aligners Sheet | 5.6 | docs/aligners/RippleLogic_Aligners_Sheet_v5.6.xlsx |
- Run-record schema v4:
schemas/mathgov_run_record_v4.schema.json - Run validator:
release/VALIDATE_MATHGOV_RUN.py - RLS Validation Workbook v0.3:
docs/validation/rls/RLS_Validation_Workbook_v0_3.xlsx - State, transition, and audit-flag registries:
docs/implementation/ - Machine vectors:
tests/ - Reproducibility, integrity, and hash checks:
release/
MHIOS v0.8 is distributed separately as a pre-1.0 candidate human-interface and orchestration companion. It is not a Core 15 governing source and does not add a cascade stage. If MHIOS conflicts with the RippleLogic Canon, the Canon controls.
Install the release dependencies from requirements.txt, then run:
python release/VERIFY_RELEASE.pyPassing the included checks establishes package, schema, formula, rendering, and tested-interface conformance. It does not prove evidence truth, empirical effectiveness, physical safety, legal authority, consciousness, moral truth, framework superiority, or deployment readiness.
- MHIOS — MathGov Human Interface and Orchestration Standard
Operationalizes the human-interface, orchestration, authority, computational-context, and execution-control layer around MathGov / RippleLogic decisions.
https://github.com/MathGov/mhios
MHIOS does not redefine the RippleLogic Canon. It provides a distinct implementation and orchestration layer synchronized with the current MathGov Core baseline.