TC56 on Natural Language Interaction Protocol (NLIP) for communication with AI agents call for participation and request for comments
Ecma TC56 committee is calling for participation and for feedback from interested stakeholders on the development of a standard protocol for communication among AI agents.
Background
Generative AI (GenAI) is reshaping the digital landscape by performing complex tasks once thought to be uniquely human—from translating languages and writing code to generating rich, multi-modal content like images, audio, and video. As innovation accelerates, a new frontier in AI is emerging: agentic AI.
Agentic AI refers to systems of digital agents that combine multiple GenAI capabilities to evaluate problems, explore solutions, and take proactive actions on behalf of humans or other AI agents. In enterprise settings, these agents are gaining advanced abilities in contextual reasoning, complex software navigation, information retrieval, question answering, fact-checking, data-driven decision-making, task planning, and problem-solving.
Driven by real-world objectives, an AI agent may coordinate with other agents and tools to solve problems across specialized domains. These systems can support a wide range of use cases—including diagnosing and resolving high-priority IT outages, resetting user passwords, coordinating enterprise-wide responses to phishing attacks, onboarding new employees, managing product returns, supporting software testing, or preparing financial reports for executive review.
Just as the web browser unlocked a new era of connected digital services, agentic AI promises to do the same for knowledge work by connecting services, users, and organizations across boundaries.
However, this evolution requires thoughtful design to ensure safety, trust, and ethical deployment. A secure, cross-platform communications protocol is essential to prevent ecosystem fragmentation from incompatible or proprietary solutions. Ecma TC56’s NLIP (Natural Language Interaction Protocol) addresses this need with an open, interoperable standard for agent communication—designed from the ground up for simplicity, scalability, and security, NLIP enables seamless interaction between humans and AI agents.
The development of this protocol began in March 2024 through a collaborative initiative within the Enterprise Neurosystems Group (ENG), an open-source consortium. This effort led to the formation of TC56 within Ecma International in December 2024. On 1 May 2025, TC56 approved the first draft specification of the NLIP protocol—marking a major milestone in building a more open, secure, and intelligent foundation for the agentic AI era. NLIP offers a path toward global alignment, helping industry, governments, and developers converge on shared standards for safe and interoperable AI agent communication.
Timeline for expression of participation
Interested stakeholders are kindly invited to ask questions, express their comments and provide feedback on the proposed first draft of the standard protocol for agent-to-agent communication.
The committee is publishing the first draft specification to enable the development of royalty-free open-source proof-of-concept reference implementations and will incorporate validated feedback from those implementations and community comments, with the second draft to be published around 1 July 2025. Comments are requested before 20 June 2025 so they can be discussed and incorporated into the second draft specifications.
TC56 expects to submit the final draft specifications for approval to the Ecma General Assembly in December 2025, after completing the aforementioned iterations of the draft specifications.
Ecma members can download the specification and comment on the Ecma members site. For the convenience of stakeholders who are not Ecma members, the draft specifications is available for review on GitHub to solicit comments.
More details can be found in the call for participation document.