aetre

# AETRE & The Governed Agent: Unified Decision-Theoretic Operating System

[![Zenodo DOI (AETRE software)](https://zenodo.org/badge/1346232534.svg)](https://doi.org/10.5281/zenodo.22098366)
[![Zenodo DOI (replication bundle)](https://zenodo.org/badge/DOI/10.5281/zenodo.22814799.svg)](https://doi.org/10.5281/zenodo.22814799)
[![SSRN: 7161458](https://img.shields.io/badge/SSRN-7161458-blue.svg)](https://ssrn.com/abstract=7161458)
[![License: AGPL v3](https://img.shields.io/badge/License-AGPL%20v3-green.svg)](LICENSE)
[![Rust: 1.75+](https://img.shields.io/badge/Rust-1.75%2B-orange.svg)](https://www.rust-lang.org/)
[![Model Context Protocol](https://img.shields.io/badge/MCP-24%20Tools-purple.svg)](https://modelcontextprotocol.io/)
[![Live Portal](https://img.shields.io/badge/Portal-lithiumeel.com%2Faetre-emerald.svg)](https://www.lithiumeel.com/aetre)

> **"Governing Autonomous Intelligence in the Age of Abundance."**  
> A high-performance, mathematically unified decision-theoretic operating system uniting macroeconomic proposal/portfolio triage (AETRE) and microeconomic execution governance (The Governed Agent).

> **Release status: experimental public alpha.** The software and mathematical
> simulations are testable, but the bundled data are synthetic and do not
> establish prospective effectiveness in a live conference, grant, or
> investment workflow. Use outputs as decision-support diagnostics, not as
> autonomous acceptance, rejection, funding, or investment decisions.

Based on the research series by Clayton Gray (2026):  
1. **The Innovation-Absorption Gap: How Artificial Intelligence Can Accelerate Idea Production Faster Than Complementary Institutions Adapt**  
   *Clayton Gray (2026a)* — [SSRN: 7161458](https://ssrn.com/abstract=7161458)
2. **The Admission Frontier: An Economically Regulated Decision-Theoretic Runtime and Fail-Closed Verification Gate for Autonomous Agents**  
   *Clayton Gray (2026d)* — distributed within the replication bundle *The Implementation Frontier: Capital Allocation, Deliberative Stopping, and Verified Agent Gatekeeping* ([Zenodo: 10.5281/zenodo.22814799](https://doi.org/10.5281/zenodo.22814799))

This software is archived under its own DOI, separate from the papers above:
**AETRE: Adaptive Epistemic Triage & Recall Engine** ([Zenodo: 10.5281/zenodo.22098366](https://doi.org/10.5281/zenodo.22098366)).

---

## The Problem: The Innovation-Absorption Gap

When Artificial Intelligence makes idea and action generation cheap ($c_{\text{gen}} \to 0$), proposal and execution volume ($N$) explodes. However, downstream evaluation, laboratory validation, code review, and human gatekeeper capacity ($K$) remain strictly finite.

This creates three critical pipeline pathologies:
1. **The Kingman Delay Explosion:** When evaluator utilization $\rho = \lambda / \mu$ approaches saturation ($\rho > 0.85$), wait times shoot up non-linearly according to Kingman's Heavy-Traffic equation:
   $$E[W_q] \approx \frac{\rho}{1-\rho} \cdot \frac{c_a^2 + c_s^2}{2} \cdot \frac{1}{\mu}$$
2. **The Asymmetric Payoff Trap:** In heavy-tailed domains like venture capital, breakthrough discovery, and agentic code patches (Pareto index $\alpha \approx 1.25$), consensus-seeking scoring systems penalize high-variance, transformative outliers in favor of safe, incremental proposals.
3. **The Finite-Capacity Recall Ceiling (Proposition 1):** Without active epistemic triage, true breakthrough recall asymptotically decays towards zero as arrival rates surge:
   $$R_N \le \min\left(1, \frac{K_N}{H_N}\right) \to 0 \quad \text{as } N \to \infty$$

---

## Two-Layer Decision-Theoretic Architecture

AETRE unifies **Macroeconomic Pipeline Triage** (managing institutional review bandwidth and portfolio recall) with **Microeconomic Execution Governance** (safeguarding agentic execution runtimes and pull-request verification).

```text
========================================================================================
 MACRO LEVEL: Institutional Proposal & Portfolio Triage (AETRE / Gray 2026a)
========================================================================================
                     Incoming Submissions / Dealflow (N)
                                     │
                                     ▼
        ┌─────────────────────────────────────────────────────────┐
        │ 1. Bayesian Value-of-Information (VOI) Engine           │
        │    - Closed-form normal VOI & Pareto heavy-tailed VOI   │
        │    - Direct-Pass, Fast-Drop, or Deep-Review allocation  │
        └─────────────────────────────────────────────────────────┘
                                     │
                                     ▼
        ┌─────────────────────────────────────────────────────────┐
        │ 2. Kingman Heavy-Traffic Capacity Governor              │
        │    - Dynamic queue throttling when ρ → 1.0              │
        │    - Preserves reviewer quality; deters burnout         │
        └─────────────────────────────────────────────────────────┘
                  │                                     │
                  ▼                                     ▼
        [ Admitted Cohort (K) ]             [ Exploration Audit Pool ]
        Optimal Conviction Allocation       Horvitz-Thompson H_hat_D Unbiased Audit

========================================================================================
 MICRO LEVEL: Autonomous Execution & Verification Gate (Governed Agent / Gray 2026d)
========================================================================================
                     Autonomous PRs / Candidate Actions
                                     │
                                     ▼
        ┌─────────────────────────────────────────────────────────┐
        │ 3. Tripartite Review Boundary & Bellman DP (Road A)     │
        │    - Computes Expected Value of Verification (V*)       │
        │    - Classifies AUTO_EXECUTE, REQUIRE_REVIEW, or REJECT │
        │    - Fail-Closed: Rejects on malformed input/divergence │
        └─────────────────────────────────────────────────────────┘
                                     │
                                     ▼
        ┌─────────────────────────────────────────────────────────┐
        │ 4. Heterogeneous cμ-Rule Knapsack Controller            │
        │    - Value density sorting: ρ_i = Δu_i / c_i            │
        │    - Budget-constrained admission under shadow prices   │
        └─────────────────────────────────────────────────────────┘
                  │                                     │
                  ▼                                     ▼
        [ Auto-Merged / Dispatched ]        [ Escalated to Human Gatekeeper ]
```

---

## Repository Structure

```text
.
├── Cargo.toml                  # Workspace manifest (AGPL-3.0)
├── crates/
│   ├── aetre-core/             # Pure Rust decision engine:
│   │   ├── src/voi.rs          #   - Bayesian VOI & heavy-tailed Pareto
│   │   ├── src/governor.rs     #   - Bellman DP & Tripartite Review Boundary (Road A)
│   │   ├── src/knapsack.rs     #   - Heterogeneous cμ-rule knapsack controller
│   │   ├── src/queues.rs       #   - Kingman heavy-traffic capacity governor
│   │   ├── src/audit.rs        #   - Horvitz-Thompson exploration audit
│   │   └── src/staking.rs      #   - Super-linear anti-sybil staking
│   ├── aetre-cli/              # Native CLI for simulations, bounds & backtests
│   └── aetre-mcp/              # Model Context Protocol server (24 tools, 4 resources, 3 prompts)
├── python/
│   └── governed_agent/         # Python reference runtime for agent execution governance
├── scripts/
│   └── pre-commit-governed-gate.py  # Standalone pre-commit verification gatekeeper
├── tests/                      # Python unit & regression suite (79 tests)
├── benchmarks/                 # Verification benchmarks and institutional queue sweeps
├── examples/
│   ├── datasets/               # Held-out review and dealflow test splits
│   ├── proposals.json          # Benchmark evaluation candidates
│   └── mcp_config.json         # Claude Desktop & Cursor connection template
├── .pre-commit-hooks.yaml      # Pre-commit hook definition for git integration
├── .github/workflows/
│   ├── ci.yml                  # Rust & MCP server automated verification
│   └── governed-gate-template.yml # Reusable GitHub Actions agent PR gating workflow
├── CITATION.cff                # Dual academic citation metadata
├── Dockerfile                  # Production container definition
├── fly.toml                    # Serverless Cloud deployment config
├── DATASETS.md                 # Fixture provenance and third-party data guidance
├── LICENSE                     # GNU Affero General Public License v3.0 text
├── LICENSING.md                # AGPL/commercial licensing overview
└── README.md
```

---

## Quickstart & CLI Usage

### 1. Run the Rust Test Suite & Verification
```bash
cargo test --workspace
cargo clippy --workspace --all-targets -- -D warnings
```

### 2. Run the Python Reference Governance Suite
```bash
python -m unittest discover -s tests
```

### 3. Run the Standalone Governed Agent Pre-Commit Gate
Fast, fail-closed verification gate for agentic code modifications:
```bash
python scripts/pre-commit-governed-gate.py --all-files
```

### 4. Run the Macro Monte Carlo & Dealflow Benchmarks
```bash
# Multi-regime academic triage simulation (500 replications)
cargo run -p aetre-cli -- benchmark --replications 500

# Venture Capital Pareto dealflow benchmark (10,000 deals, α = 1.25)
cargo run -p aetre-cli -- vc-benchmark --deals 10000 --budget 100 --alpha 1.25

# Theoretical Proposition 1 recall ceiling bound
cargo run -p aetre-cli -- bound --arrivals 5000 --capacity 200 --high-rate 0.067 --csv
```

### 5. Simulate Institutional Heterogeneous Agent Queues
Simulates density-greedy $c\mu$-rule knapsack vs. FIFO across varying institutional review capacities:
```bash
python evaluate_institutional_queues.