Governance Infrastructure for Autonomous Decisions

Quantitative decision governance for engineering teams. Evaluate proposals through mathematical gates — not runtime guardrails or compliance checklists.

AEGIS evaluates proposals through six quantitative gates — returning structured PROCEED / PAUSE / ESCALATE / HALT decisions with hash-chained audit trails and compliance-aligned documentation.

Get Started Try the Advisor API Reference Portal

The Problem

AI agents and engineering teams make high-stakes decisions every day — deployments, architecture changes, resource allocations. But there is no standardized way to evaluate those decisions, and no audit trail to prove they were sound.

How AEGIS Solves It

Every decision gets a unique ID, timestamp, rationale, and cryptographically verifiable audit log.

How It Works

Proposal → [Risk] [Profit] [Novelty] [Complexity] [Quality] [Utility] → Decision
                                                                          ↓
                                                          PROCEED / PAUSE / HALT / ESCALATE
                                                          + confidence + rationale + next steps

Every proposal is evaluated against six gates:

• Risk — Is the risk delta acceptable? Bayesian posterior confidence evaluation.
• Profit — Does the expected value justify the change? Bayesian confidence check.
• Novelty — How novel is this approach? Logistic function scoring.
• Complexity — Is the system complexity within bounds? Hard floor enforcement.
• Quality — Does code quality meet standards? Minimum score with no zero subscores.
• Utility — Is the net utility positive? Lower confidence bound verification.

Quick Example

from aegis_governance import AegisConfig, PCWContext, PCWPhase, pcw_decide

config = AegisConfig.default()
evaluator = config.create_gate_evaluator()

decision = pcw_decide(
    PCWContext(
        agent_id="my-agent",
        session_id="session-1",
        phase=PCWPhase.PLAN,
        proposal_summary="Add Redis caching layer",
        estimated_impact="medium",
        risk_proposed=0.15,
        complexity_score=0.7,
    ),
    gate_evaluator=evaluator,
)

print(decision.status.value)  # "proceed", "pause", "halt", or "escalate"

curl -X POST https://aegis-api-980022636831.us-central1.run.app/evaluate \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "proposal_summary": "Add Redis caching layer",
    "estimated_impact": "medium",
    "risk_proposed": 0.15,
    "complexity_score": 0.7
  }'

echo '{"proposal_summary":"Add caching","estimated_impact":"low"}' \
  | aegis evaluate

- uses: undercurrentai/aegis-governance/.github/actions/aegis-gate@main
  with:
    proposal_summary: "Add Redis caching layer"
    estimated_impact: medium
    risk_proposed: 0.15

Integration Options

How AEGIS Is Different

Most approaches to AI governance fall into two categories: runtime agent guardrails that monitor and gate what agents can do in real-time, or compliance dashboards that manage policies through checklists and binary pass/fail scoring. Both have a blind spot — neither evaluates whether the engineering decision itself is mathematically sound.

AEGIS takes a third approach: quantitative decision governance. Instead of watching agent behavior or managing compliance checklists, AEGIS evaluates the proposal through six mathematical gates — Bayesian posteriors, logistic functions, utility theory, and hard-floor enforcement. The question isn't "can this agent act?" or "does this check a compliance box?" — it's "should this specific engineering change ship?"

Key Features

• Hash-chained audit trails — tamper-evident decision logs for compliance and accountability
• Enterprise-grade security — hybrid signatures combining classical (Ed25519) with post-quantum (ML-DSA-44, FIPS 204) and encryption (ML-KEM-768, FIPS 203). The first governance platform with quantum-resistant audit trails, prepared for CNSA 2.0 timelines.
• Risk-free rollout — shadow mode lets you evaluate without affecting production decisions (Professional tier and above)
• Continuous monitoring — drift detection alerts when decision patterns shift from baselines
• AI-agent native — MCP server, GitHub Action, and SDK for autonomous agent governance
• Compliance-aligned — artifacts aligned with NIST AI RMF, EU AI Act, and SOC 2 trust criteria
• Configurable thresholds — YAML-driven parameter management with version-controlled schemas

Built for Regulated Environments

AEGIS includes compliance artifacts and runbook templates aligned with these frameworks. Alignment artifacts, not certifications. See compliance maturity.

View compliance documentation