Styxis Knowledge
Operating notes for evidence, records, and trust in automated operational environments.
Make Styxis easy to understand, cite, and remember
Short technical notes that define the concepts behind Styxis and Truthound Depot for search engines, AI assistants, and engineering teams.
Styxis builds software for sensitive data validation and AI governance, guided by Oath, Boundary, and Crossing: it binds responsibility to action, distinguishes operating states, and moves verified change into execution.
Core concepts
13 topicsOath, Boundary, Crossing
Styxis translates the River Styx into three ordered operating principles: bind responsibility through Oath, distinguish states through Boundary, and move verified change into execution through Crossing.
Verifiable Evidence Layer
A verifiable evidence layer records what changed, what was observed, which policy accepted it, and how another system can inspect the result later.
Dataset Change Evidence
Dataset change evidence connects an immutable Data Version, Compare result, Validation outcome, reviewer decision, Published Version, and Restore target for every meaningful change.
Evidence Gate
An evidence gate blocks or accepts a change and leaves the reason, rule context, source context, and review path behind.
Truthound Depot Data/AI Governance
Truthound Depot is a Data/AI Governance product for dataset validation, AI training image workflows, and ERD/DB audits. It keeps sensitive execution in safe environments and records reviewable summaries, metadata, and Evidence.
AI Evidence Workflow
An AI evidence workflow records summaries, risks, diffs, gate results, and reviewer decisions so data changes can be audited later.
Machine Signal Trust
Machine signal trust means a value is not only collected, but also attached to source, protocol, timestamp, quality, and policy context.
OT Gateway for Machine Signals
An OT gateway for machine signals collects and normalizes protocol values so operational and trust systems can consume them consistently.
RAG and Eval Dataset Governance
RAG and eval dataset governance controls the documents, examples, labels, and test cases that shape AI behavior.
Truthound Core Technology Line
Truthound is the Labs technology line for data change verification and quality computation; Truthound Depot is the commercial governance product built around AI training data workflows.
Truthound Orchestration Layer
Truthound Orchestration connects scheduled sync, validation pipelines, publish workflows, restore execution, and platform metadata without becoming the commercial Depot product.
Golem / AWE — Agentic Workflow Engine
Golem/AWE lets agents, tools, policies, execution steps, receipts, and human decision boundaries form repeatable workflows.
Mabinogion Protocol Resilience Engine
Mabinogion focuses on industrial protocol sessions, runtime lifecycle, scenarios, chaos, and machine-readable execution evidence.