Augury, a leading Industrial AI company, has introduced the OKF Industrial Profile, an open-source manufacturing-specific extension of Google Cloud’s Open Knowledge Format (OKF). Created to simplify the complexities of industrial operations, the profile offers a vendor-neutral framework for organizing plant-floor context, operator expertise, and sensor-trust policies into human-validated and auditable ground truth that can be effectively used by AI agents.
As AI agents are introduced to manufacturing workflows, they face a critical challenge: context fragmentation. Decades of critical operational wisdom- how a process line behaves during cold starts, how an experienced lead handles shift handovers, or when a sensor reading shouldn’t be trusted remain unwritten or locked in legacy silos.
Google Cloud introduced OKF in June 2026 as an open specification for structuring knowledge as lightweight, human- and machine-readable files. Augury is releasing an industrial profile built on top of unmodified OKF, adding the specific conventions manufacturing needs to bring human judgment and operational context into AI reasoning.
“Industrial AI cannot succeed on generic software code or raw machine telemetry alone; it requires real operational context that is human-verified. The true bottleneck in plant-floor AI isn’t model intelligence; it’s capturing the unwritten wisdom of expert workers before they retire. The OKF Industrial Profile allows manufacturers to structure human experience, cause-and-effect relationships, and sensor-trust policies into version-controlled, auditable files. It gives AI agents the context they need to know when to act, and just as importantly, when not to act.”
–Anoop Mohan, Chief Product and Technology Officer of Augury.
A Human-Reviewed & Verified Layer That Complements Plant Systems
Existing industrial standards excel at what they were built for: ISA-95 for plant hierarchies and enterprise-to-control integration, ISA-88 for batch procedures and recipe models, and the Asset Administration Shell (AAS) for structured digital-twin data, while modern data lakehouses govern data pipelines and semantic data modeling.
The OKF Industrial Profile operates alongside these foundations to solve a distinct problem: standardizing human judgment, provenance, and operational trust rules. Crucially, the profile dynamically references operating limits, setpoints, and live streams directly from their source systems without duplicating values, ensuring operational knowledge never drifts out of sync with physical assets.
The OKF Industrial Profile introduces standard conventions for:
- Human Operational Knowledge with Provenance: Operator heuristics captured with source receipts, verbatim quotes, contributor credentials, and human review status.
- Data-Trust Policies: Explicit rules instructing AI agents how to handle stale or stuck telemetry feeds before making diagnostic recommendations.
- Operational Relationships: Distinguishing actual cause-and-effect drivers from duplicate or redundant sensor signals.
- Standards Mapping: An open convention that maps concepts to identifiers from existing plant-floor standards, such as ISA-95 for equipment or ISA-88 for procedures, and works alongside whatever standards a site already uses.
Building on Augury’s Industrial AI Workforce
The OKF Industrial Profile aligns directly with Augury’s Industrial AI Workforce vision: bringing together machine data, operational context, and human expertise so AI agents can work safely alongside plant personnel.
By establishing an open, human-readable, and verifiable profile, industrial organizations can structure their know-how once and deploy it across any AI model, framework, or cloud—even remaining fully functional at disconnected edge plant sites.
“When we introduced the Open Knowledge Format, our goal was to establish a vendor-neutral standard for structuring context across enterprise workflows. Augury’s contribution to the industrial profile demonstrates the power of open ecosystems. By unifying human operational expertise, sensor trust rules, and physical context within the OKF, this initiative enables AI agents to deliver safer, more reliable outcomes in manufacturing environments.”
–Praveen Rao, Global Head of Manufacturing Industry at Google Cloud.
Open Contribution and Availability
To support broad community adoption, Augury is making the OKF Industrial Profile available today in a public repository under the Apache 2.0 license. Augury also intends to contribute the showcase bundle to the upstream OKF open-source project, following the project’s contribution process and the pattern of its existing showcase bundles.
The open release includes:
- An industrial showcase knowledge bundle representing common plant assets and their operational relationships.
- The OKF Industrial Profile conventions and vocabulary documentation.
“In manufacturing, context is physical, mechanical, and deeply tied to worker experience. We are contributing this profile openly to the ecosystem because industrial AI shouldn’t be locked inside proprietary walled gardens; open conventions are the only way to achieve true interoperability across the plant floor.”
– Anoop Mohan, Chief Product and Technology Officer of Augury.
A leading Industrial AI company, Augury helps the world’s manufacturers leverage real-time production insights and data to drive new levels of efficiency. By combining predictive and prescriptive technology with operational data and customized AI agents, production teams can improve uptime, reduce costs, and maximize yield and capacity. Our mission is to transform how people and machines work together to push the boundaries of human productivity.
