Operate data and AI work that needs validation without taking the sensitive material outside.
Based in Jeju, South Korea, Styxis helps teams operate sensitive data validation, AI training data workflows, and reviewable execution without moving private material out of its safe boundary.
Our primary public product, Truthound Depot, is a Data/AI Governance product for dataset validation, AI training image workflows, and ERD/DB audits. It offers an online demo for non-sensitive data and scopes real operations through customer-hosted installation.
As an oath on the Styx bound words to consequence, Styxis connects every AI and system action to its goal, context, authority, policy, and decision maker. What ran and why must remain clear.
As the Styx marks a boundary between worlds, Styxis distinguishes observations from AI interpretation, permitted from prohibited actions, and proposals under review from approved execution.
Styxis turns proposals into data changes, releases, approvals, and operational actions only after their goals, authority, and required reviews are confirmed. Outcomes remain inspectable and recoverable.
Styxis focuses on safe execution boundaries, repeatable validation, and Evidence that lets teams approve data and AI work from the same facts.
01 · OT Gateway for Machine SignalsNo Protocol Escapes.
A Rust-based OT gateway and protocol runtime that collects and normalizes Modbus TCP/RTU and OPC UA industrial protocol signals.
02 · Data/AI GovernanceSensitive data stays put. Reviewable Evidence moves forward.
Validate datasets, AI training images, and ERD/DB implementation inside safe execution boundaries. Try the online demo with non-sensitive data, then scope real operations through a customer-hosted installation.
"No Protocol Escapes."
A Rust-based OT gateway that brings industrial protocol signals into the Styxis trust layer.
Collect and normalize machine signals from Modbus, OPC UA, and more.
"Keep sensitive data in place. Leave verifiable results behind."
Truthound Depot processes dataset validation, AI training image workflows, and ERD/DB audits in safe execution environments.
It is not a product for exporting original data, database credentials, or private images. It records the summaries, metadata, and Evidence teams need to review, repeat, and approve work.
Inspect dataset versions for PII, validation rules, schema drift, anomalies, and review impact before teams approve the change.
Review image sets, annotations, quality signals, and generation workflow metadata without exposing private source images as product data.
Audit ERDs against standard words, terms, and domains, then verify whether the approved design was implemented in the physical database.
Truthound Depot gives reviewers the facts they need without turning private data, credentials, or non-public images into exported product payloads.
Version datasets, run PII, validation, schema, drift, and anomaly checks, and compare results before a change is approved.
Inspect training image quality, annotation consistency, and generation workflow metadata while private image assets stay inside the agreed boundary.
Run validation close to the source with local credentials and read-only access, then return reviewable result contracts instead of raw private data.
Check whether ERD design follows approved standard words, standard terms, domain definitions, naming rules, and relationship intent.
Verify that the actual database implements the approved ERD with matching physical names, logical names, tables, columns, keys, and structure.
Record summaries, metadata, issue context, reviewer decisions, and reports so repeat review and approval can happen from the same basis.
From shared working agreements to data changes, Styxis makes important work traceable, reviewable, and verifiable.
Software Sworn to Truth
