Repository navigation
Releases
AetherGovern v2.0.0 — Reference Implementation of the Falsification-First Governance Architecture
AetherGovern v2.0.0 — Reference Implementation
Status: Reference Implementation (preliminary results). The v0.1.0 hypothesis (F1–F5) is not falsified by the implementation or the experiments, but it is also not confirmed.
This is the software release (v2.0.0) of the AetherGovern falsification-first governance architecture specified in v0.1.0. The v0.1.0 paper remains the authoritative hypothesis record; v2.0.0 adds a reference implementation alongside it.
🔗 Links
- 📦 PyPI package: aethergovern 2.0.0
- 🐙 GitHub repository: gitdeeper15/aethergovern
- 📄 Paper (v0.1.0): 10.5281/zenodo.24054524
- 🌐 Website: aethergovern.netlify.app
- 📋 OSF Registration: 10.17605/OSF.IO/TAHC8
- 🔗 ORCID: 0009-0003-8903-0029
What v2.0.0 adds to v0.1.0
| v0.1.0 (hypothesis) | v2.0.0 (this release) |
|---|---|
| Formal architecture specified | Reference implementation built |
| Two propositions (proof sketches) | Implemented causal-authority admission and monotonic revocation |
| Five failure conditions (F1–F5) | Evaluated against M1–M5 (simulation) |
| No implementation, no tests, no experiments | 15 modules, 60 tests, 3 experiments |
| No PyPI package | aethergovern 2.0.0 on PyPI |
Implementation status
| Component | Status |
|---|---|
| Modules | 15 (Python, ~1,540 lines) |
| CLI | aethergovern info · aethergovern verify |
| Tests | 60 passed (pytest) |
| Experiments | 3/3 completed (simulated data) |
| PyPI | aethergovern 2.0.0 — Published |
Experimental results (simulation only)
| # | Experiment | Result |
|---|---|---|
| 1 | Authorization Correctness (M1) | 0.0000 unauthorized admission rate (10–500 nodes) |
| 2 | Exposure Window (M2, M3) | M2 = 0; M3 median = 100 ms |
| 3 | Attribution Accuracy (M4, M5) | M4 = 1.0; M5 = 1.0 |
Caveat: Experiments use simulated data. These are preliminary results; no real-world deployment has been evaluated.
What v2.0.0 does not change
- The v0.1.0 hypothesis (F1–F5) — still the authoritative record.
- Propositions P1–P2 — still proof sketches, not machine-checked.
- Real-world deployment — still not evaluated.
Installation
pip install aethergovernRequires Python 3.9+. Runtime dependency: cryptography >= 41.0.
Known limitations
- Simulation only. No real-world deployment.
- Simplified network model.
- Simplified Byzantine model.
- Root registry is centralized.
- No ablation.
- No mechanized proofs (P1–P2 remain proof sketches).
- No performance benchmarks.
"AetherGovern is a candidate, not a result."
— AetherGovern v0.1.0, Section 8.5
AetherGovern v0.1.0 — A Falsification-First Investigation of Governance in the Aether Family
AetherGovern v0.1.0 — Architectural Hypothesis Under Investigation
Status: Architectural Hypothesis Under Investigation (v0.1.0). This is not a committed project. It is a structured investigation of whether a governance gap genuinely exists in the Aether family. No implementation is claimed. No real-world deployment has been evaluated. The final project structure (single, dual, or triple project) will be determined after comparison.
Abstract
The Aether family comprises four preprints that address, respectively, how agent-generated changes are stored (AetherGit), integrated and verified operationally (AetherFlow), specified and formally evidenced (AetherProof), and synchronized without a central coordinator (AetherMesh). None of them is primarily about authority: who may act, under which policy, with what accountability, and how authority is withdrawn. This paper investigates whether that omission is a genuine gap that warrants a fifth functional dimension, tentatively named AetherGovern, or merely an artifact of how the existing dimensions were scoped.
It does so with a falsification-first method: before describing the candidate, it states five failure conditions (F1–F5) whose satisfaction would show that the candidate does not deserve to exist as a separate dimension, together with a decision rule. It then compares three candidate capabilities (orchestration, governance, and runtime assurance), specifies a five-layer governance architecture (authorization, policy, accountability, revocation, dispute resolution), and states two propositions with proof sketches: that authority checks which cite their own revocation frontier preserve the determinism that AetherMesh's convergence argument requires, and that revocation in an eventually consistent setting leaves an exposure window that no decentralized design can remove.
On the basis of argument alone, conditions F1–F4 are provisionally not falsified, with stated residual weaknesses, and F5 (observable benefit) is untested. The paper therefore concludes that AetherGovern is a candidate, not a confirmed dimension. No implementation, experiment, or measurement is reported.
Keywords: Governance; Authorization; Delegation; Revocation; Accountability; Multi-Agent Systems; Falsification; Decentralized Systems; Agentic Software Engineering; AetherGovern.
What's in this release
Manuscript only.
- Research Paper (PDF) — 33 pages
- 4 formal definitions (as layers)
- 2 propositions (with proof sketches)
- 5 failure conditions (F1–F5)
- 6 requirements (R1–R6)
- 20 open questions (Q1–Q20)
- 25 references
- 5 figures (Fig1–Fig5)
- 4 fifth-layer-study markdown files (Orchestrate, Govern, Runtime, Comparison)
Published
- Zenodo DOI: 10.5281/zenodo.24054524
- Concept DOI: 10.5281/zenodo.24054523
- OSF Registration: 10.17605/OSF.IO/TAHC8
- Website: aethergovern.netlify.app
Research Status Statement
| Item | Status |
|---|---|
| Document type | Research article: structured investigation of an architectural hypothesis |
| Version / record | 0.1.0, November 2026 · DOI 10.5281/zenodo.24054524 |
| Project status | Not a committed project; final structure (single, dual, or triple) undetermined |
| Implementation | None claimed |
| Real-world deployment | None; none evaluated |
| Empirical results | None reported |
| Propositions 1–2 | Stated with proof sketches; not machine-checked |
| F1–F4 | Not falsified by argument alone, with residual weaknesses; provisional |
| F5 | Untested |
| Review status | Not peer reviewed |
Evidentiary status
Every statement in this paper is labeled as a definition, a hypothesis, or an observation. Propositions are stated with proof sketches and are not machine-checked. No empirical result is reported. Descriptions of third-party work follow the cited sources and were not independently verified.
The Five Functional Dimensions
| Dimension | Focus | Question |
|---|---|---|
| AetherGit | Storage and versioning | How are changes stored and versioned? |
| AetherFlow | Integration and operational verification | How are changes integrated and verified? |
| AetherProof | Specification and formal evidence | How is correctness proven? |
| AetherMesh | Distribution and eventual consistency | How is state synchronized without a central coordinator? |
| AetherGovern (hypothesis) | Authority and policy | Who is authorized to do what, under which policy? |
The Central Question
Is there a body of authorization, policy, accountability, revocation, and dispute-resolution requirements for multi-agent software ecosystems that (i) is not already specified by the four existing dimensions, (ii) is not reducible to verification or to orchestration, (iii) can be specified without reintroducing a central authority, and (iv) yields an observable benefit over existing mechanisms?
Failure Conditions F1–F5
| ID | Claim under test | Status |
|---|---|---|
| F1 | Non-redundancy | Not falsified (R1, R3, R5) |
| F2 | Orthogonality to verification | Not falsified (narrowly) |
| F3 | Boundary with orchestration | Not falsified (decentralized setting) |
| F4 | Decentralizability | Not falsified (with residuals) |
| F5 | Observable benefit | Untested |
A Modest Conclusion
"AetherGovern is a candidate, not a result."
— AetherGovern v0.1.0, Section 8.5
Citation:
@techreport{baladi2026aethergovern,
author = {Baladi, Samir},
title = {AetherGovern: A Falsification-First Investigation of
Governance in the Aether Family},
institution = {Ronin Institute / Rite of Renaissance},
type = {Research Paper},
year = {2026},
month = {November},
version = {0.1.0},
doi = {10.5281/zenodo.24054524},
url = {https://doi.org/10.5281/zenodo.24054524},
note = {Architectural Hypothesis Under Investigation — Not a committed project.}
}