Hosted service offline; see undercurrentholdings.com/status

AEGIS
Documentation

AEGIS evaluates engineering proposals through six gate checks over caller-supplied metrics and records each authenticated decision in a hash-chained audit trail. The hosted API and MCP endpoint are offline. The Python client, the local MCP server, and the offline attestation verifier are published on PyPI with their source.

Get StartedAPI Reference (not served)Advisor (questionnaire)Product Status

How It Works

Every proposal is evaluated against six gate checks over metrics the caller supplies. The result is a structured decision — PROCEED, PAUSE, HALT, or ESCALATE — with confidence scores, rationale, and next steps. Every authenticated evaluation is hash-chained; sandbox evaluations are not recorded.

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

Integration Options

Python SDK
Client of the hosted API (offline)
pip install aegis-governance
REST API
Reference only; not served
POST /evaluate
CLI
In-process engine by default (private repository)
aegis evaluate
MCP Server
Local stdio; tools call the hosted API (offline)
aegis-mcp-server
GitHub Action
Calls the hosted API (offline)
aegis-gate@main

Key Features

Six Gate Checks
Risk, profit, novelty, complexity, quality, utility, over metrics the caller supplies: Bayesian posteriors, logistic functions, and hard-floor enforcement against documented thresholds. All six must pass.
Hash-Chained Audit Trails
Every evaluation gets a unique ID, timestamp, gate results, rationale, and a SHA-256 chain entry that can be verified against the stored chain. The chain detects tail deletion and entry edits; it is not independently anchored.
Hybrid Post-Quantum Cryptography
Signatures combining classical (Ed25519) with post-quantum (ML-DSA-65, FIPS 204), and encryption with ML-KEM-768 (FIPS 203).
Shadow Mode and Drift Detection
Evaluate without enforcing the decision. KL-divergence comparison of decision patterns against a baseline.
Configurable Thresholds
YAML-driven parameter management with version-controlled schemas. Domain templates for CI/CD, finance, healthcare, and infrastructure.

Governance Templates

The repository contains self-authored governance templates and runbooks. They are not certifications, and no third-party compliance audit or certification assessment has been conducted on AEGIS itself. See compliance maturity →