IndustriConnect: MCP Adapters and Mock-First Evaluation for AI-Assisted Industrial Operations
Melwin Xavier, Melveena Jolly, Vaisakh M A, Midhun Xavier · 2026
Read IndustriConnect: MCP Adapters and Mock-First Evaluation for AI-Assisted Industrial Operations by Melwin Xavier, Melveena Jolly, Vaisakh M A, Midhun Xavier (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-24703.txt.
Introduction
Xavier and co-authors present IndustriConnect, a prototype suite of adapters linking AI assistants to plant systems. Assistants cannot natively speak Modbus, MQTT/Sparkplug B, or OPC UA. Their adapters expose plant operations as tools an AI can find by schema, while keeping protocol-specific safety controls. A mock-first workflow lets adapter behaviour be tested locally before plant connection. A benchmark ran 870 trials across normal, fault, stress, and recovery cases. The normal suite passed fully, faults produced structured errors, and recovery worked for all three protocols.
Catalog record
| CREATOR | Melwin Xavier, Melveena Jolly, Vaisakh M A, Midhun Xavier |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 36050 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.24703
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
IndustriConnect: MCP Adapters and Mock-First Evaluation for AI-Assisted Industrial Operations Melwin Xavier, Melveena Jolly, Vaisakh M A, Midhun Xavier Abstract AI assistants can decompose multi-step workflows, but they do not natively speak industrial protocols such as Modbus, MQTT/Sparkplug B, or OPC UA. This paper presents IndustriConnect , a prototype suite of Model Context Protocol (MCP) adapters that expose industrial operations as schema-discoverable AI tools while preserving protocol-specific connectivity and safety controls. The system uses a common response envelope and a mock-first workflow so adapter behavior can be exercised locally before connecting to plant equipment. A deterministic benchmark covering normal, fault-injected, stress, and recovery scenarios evaluates the flagship adapters. The benchmark comprises 870 runs (480 normal, 210 fault-injected, 120 stress, 60 recovery trials) and 2820 tool calls across 7 fault scenarios and 12 stress scenarios. The normal suite achieved full success, the fault suite confirmed structured error handling with adapter-level uint16 range validation, and the stress suite identified concurrency boundaries. Same-session recovery after endpoint restart is demonstrated for all three protocols. The results provide evidence spanning adapter correctness, concurrency behavior, and structured error handling for AI-assisted industrial operations. I Introduction Consider a simple operator request: check a temperature excursion on an OPC UA namespace, confirm a related Modbus register block, and publish a Sparkplug B status update for downstream systems. Each subtask is routine in isolation, but the overall workflow crosses incompatible operational technology (OT) protocols whose interfaces were not designed for modern AI assistants. This protocol gap remains a practical barrier to AI-assisted industrial automation even as Industry 4.0 programs push tighter coupling between information technology and plant-floor systems [ 1 , 2 ] . Large language model (LLM) assistants work best with typed, discoverable tool interfaces rather than binary protocol frames or vendor SDKs. MCP provides a standard way for assistants to list tools, inspect schemas, and invoke operations through a common call interface [ 3 ] . MCP alone, however, does not solve OT connectivity, endpoint health, or the need to represent blocked writes and transient failures in a way that an agent can reason about safely. That gap is especially visible in brownfield environments. Real plants often begin with partial register maps, vendor-specific addressing rules, or limited access windows on production equipment. A practical integration layer therefore needs more than tool schemas: it also needs protocol adapters, explicit safety conventions, and a way to validate workflows before touching live controllers. This paper reframes IndustriConnect as a bounded systems contribution: a prototype MCP-to-OT adapter architecture with a mock-first evaluation workflow. The paper does not claim a complete industrial platform, a new LLM reasoning method, or production deployment evidence. Instead, it focuses on the flagship adapters for Modbus, MQTT/Sparkplug B, and OPC UA while retaining seven additional protocol modules as ecosystem breadth. The manuscript makes three contributions: 1. A practical MCP-to-OT adapter architecture that standardizes tool discovery and response handling across heterogeneous industrial protocols. 2. A mock-first methodology for validating adapter behavior, write guards, and restart recovery, extended with fault injection and stress testing to expose error-handling and concurrency boundaries without requiring physical devices. 3. A reproducible deterministic evaluation on Modbus, MQTT/Sparkplug B, and OPC UA comprising normal, fault-injected, stress, and recovery suites using deterministic workloads with statistical variance reporting. The evaluation uses a deterministic benchmark that exercises adapter behavior under normal operation, deliberate fault conditions, stress scenarios, and endpoint restart. II Design Goals and Architecture Heterogeneity. Industrial protocols expose different interaction models: register maps for Modbus, broker topics for MQTT, and browsable information models for OPC UA [ 4 , 5 ] . The adapter layer must translate these differences into tool schemas that remain understandable to an AI client. Schema discoverability. Each adapter registers MCP tools so a client can enumerate available operations and call them without bespoke protocol code. This keeps the assistant-facing interface uniform even when the underlying transport or addressing model differs. Write safety. The adapters expose explicit write operations, but writes remain bounded by per-server configuration such as disable flags, address validation, and structured error returns. These controls matter because an AI-facing API must represent both successful writes and intended denials in a machine-readable way. Mockability. Every flagship adapter is paired with a local mock so the full tool path can be exercised offline. This enables reproducible evaluation, debugging, and regression checks before any real controller or fieldbus is involved. MCP Client / AI Assistant Claude Desktop · mcp-manager-ui · Custom LLM app stdio / JSON-RPC 2.0 (MCP) IndustriConnect Protocol MCP Server FastMCP · asyncio · write guards {success, data, error, meta} Modbus MQTT/SB OPC UA BACnet DNP3 EtherCAT EtherNet/IP S7comm … Mock Simulator Real PLC / RTU Offline testing Plant deployment mock live Fig. 1: IndustriConnect adapter architecture. An AI assistant issues MCP tool calls through a common JSON-RPC interface. The adapter layer handles protocol translation and safety guards, returning a shared response envelope. Each protocol adapter targets either a local mock (dashed) or real plant equipment. Figure 1 shows the core pattern. An MCP-compatible assistant issues tool calls to an IndustriConnect adapter. The adapter handles protocol translation, connection management, and safety guards, then talks to either a mock endpoint or plant equipment. All benchmarked adapters return the same response envelope, {success, data, error, meta} . The success flag indicates whether the requested action achieved the expected protocol outcome, data holds the protocol-specific payload, error carries a structured failure reason, and meta records endpoint details, timings, and retry counts. This shared envelope does not erase protocol semantics; it defines a common measurement boundary and lets cross-protocol orchestration observe healthy reads, blocked writes, and reconnect behavior with the same outer structure. TABLE I: Shared response-envelope fields used across the benchmarked adapters. Field Role in the adapter contract success Reports whether the requested protocol action reached the expected outcome. data Carries the protocol-specific payload such as registers, node values, or publish metadata. error Preserves a machine-readable failure reason, including guarded denials and transient endpoint failures. meta Records wall-clock latency, endpoint details, attempts, and protocol-specific trace context. Table I is operationally important because it defines how the benchmark judges behavior. A blocked write is not treated as a generic failure if the task oracle expects a guarded denial, while a reconnect path is evaluated through the same outer structure as a healthy read. This is the main reason the paper can compare three different protocol families without pretending they expose the same payload model. The mock-first workflow complements the same architecture. Each flagship adapter is exercised against a local endpoint first, but the MCP layer remains identical when the endpoint changes. That lets an engineer develop prompts, schemas, and safety checks against a deterministic mock while preserving the same tool surface for later hardware validation. III Prototype Implementation Every protocol module follows the same skeleton: an MCP server registers tools and schemas, a protocol client wrapper translates tool calls into native operations, and a mock endpoint provides a safe development target. The benchmarked Modbus, MQTT/Sparkplug B, and OPC UA stacks all use this pattern, including the shared response envelope and environment-driven write gating. The result is not a universal industrial abstraction layer; it is a consistent adapter contract around heterogeneous protocols. At implementation level, the