arXiv

LAP: An Agent-to-Instrument Protocol for Autonomous Science

Title: LAP: A Protocol for Agent-to-Instrument Communication in Autonomous Science

Abstract

Autonomous science is transitioning from proof-of-concept demonstrations to established infrastructure. While large language model agents now handle experimental planning and self-driving laboratories perform the tasks, each system currently reconstructs the connection between reasoning agents and physical instruments from the ground up. This process is complicated by fragmented vendor SDKs and standards designed for deterministic software clients, which are ill-suited for probabilistic, goal-directed agents.

While recent interoperability protocols have addressed two of the three edges in an agentic ecosystem—Anthropic’s Model Context Protocol (MCP) for agent-to-tool interactions and Google’s Agent2Agent (A2A) for agent-to-agent communication—neither adequately models the agent-to-instrument edge. This specific interface involves operations that are stateful, safety-critical, exclusively owned, physically embodied, and yield measurements complete with units, calibration data, and uncertainty.

To bridge this gap, we introduce the Lab Agent Protocol (LAP). LAP builds upon the peer-to-peer, discovery-first, task-lifecycle structure of A2A but introduces four physical-world primitives:

  1. InstrumentCard: A signed description detailing capabilities and physical limits.
  2. Exclusive Reservation: First-class locking mechanisms for instruments and samples.
  3. Safety-Fence Handshake: Operator-confirmation tokens, cryptographically bound to specific tasks and parameters, to gate hazardous or irreversible operations.
  4. MeasurementResult Schema: A structure ensuring results are physically typed (using QUDT/UCUM), calibration-anchored, uncertainty-bearing, and reproducible by design.

The paper specifies the protocol’s roles, a six-layer architecture, JSON-RPC methods, task and safety state machines, error models, and cross-laboratory federation. We demonstrate its efficacy by walking through a closed-loop autonomous campaign end-to-end. LAP is compatible with the A2A/MCP ecosystem’s transport layer and is designed to encapsulate, rather than replace, existing device standards such as SiLA 2 and OPC-UA.


Source: arXiv Generated at: 2026-06-03 00:00:00 UTC

Related Articles

TikTok Billionaire Tops Ambani as Asia’s Second-Richest
Bloomberg

TikTok Billionaire Tops Ambani as Asia’s Second-Richest

TikTok founder surpasses Mukesh Ambani to become Asia’s second-richest person, marking a significant shift in the region...

Publishers in UK can opt out of Google AI search results
BBC News

Publishers in UK can opt out of Google AI search results

UK publishers can now opt out of Google’s AI search summaries, a CMA ruling designed to boost their bargaining power and...

Kioxia Edges Nearer Toyota’s Market Cap in Shakeup to Japan Inc.
Bloomberg

Kioxia Edges Nearer Toyota’s Market Cap in Shakeup to Japan Inc.

Kioxia’s market cap nears Toyota’s, signaling a major shift in Japan’s corporate hierarchy. This narrowing gap highlight...

Reuters

Morning Bid: Marvell, a fitting name for the latest AI darling

Reuters highlights Marvell as a top AI stock, noting its name perfectly suits its status as the newest market darling.

Financial Times

Tim Hayward: I built the Jaguar E-Type of computer keyboards

Tim Hayward compares his bespoke keyboard designs to the Jaguar E-Type. He explores high-end customization for personal ...

Financial Times

AI Labs: Zuckerberg’s $100bn gamble

Meta’s $100 billion AI investment aims to secure AI dominance, but questions remain whether sheer spending can outpace c...