An entity situated in an environment that perceives its environment and acts on it, over time, in pursuit of its goals. For a detailed discussion of agent definitions, see [[FRANKLIN96]].
Agent Interaction Protocol
A specification of communication among two or more agents that states who can say what to whom and when — for example, as message sequence diagrams [[AUML]] or information flows [[BSPL]].
-
Artifact or Tool
-
A resource [[WEBARCH]] that can be shared and used by agents to support their activities. In some multi-agent systems, agents can construct artifacts to instrument their environments [[JACAMO]].
-
Augmented Language Model
A language model augmented with abilities such as reasoning, tool use, information retrieval, or storing context across interactions. Unlike an agent, an augmented language model does not actively pursue goals and is not situated in an environment. See also [[TMLR23]] and [[ANTHROPIC24]].
+
LLM Agent or Language Agent
+
An agent that relies on an LLM to guide their internal processes and interactions with the environment, while maintaining control over how they accomplish tasks [[ANTHROPIC24]][[COALA]]. [This is the sort of agent people think about when they talk about Agentic AI.]
+
Multi-Agent System (MAS)
A system composed of agents that are situated in a shared environment and interact with one another to achieve individual or collective goals. Agents can work in collaboration, cooperation, and/or competition. A MAS can be either an open or a closed system. This report is primarily concerned with open MAS.
Situatedness
The ability of an agent to interact with its environment directly through perception and action, and to respond in a timely fashion to sensory input.
+
Tool or Artifact
+
An instrument that can be shared and used by agents to support their activities. In some multi-agent systems, agents construct artifacts to instrument their environments [[JACAMO]]. In the context of agentic AI, a tool is a functional interface to a program that a language model can invoke. Tools extend the capabilities of LLMs by enabling them to retrieve knowledge not seen during training, perform complex computations, mitigate hallucinations, and perceive or act in an environment [[TOOL]].
+
+
Web-based Tool or Web-based Artifact
+
A tool or artifact represented as a resource [[WEBARCH]] and accessible through the Web. Such tools may expose interfaces over Web or non-Web protocols—for example, a weather service exposing an HTTP API, a lamp exposing a CoAP API, or a telemetry service exposing an MQTT API. Non-Web protocols can be encapsulated behind hypermedia controls published in a description accessible through the Web, such as a W3C Web of Things (WoT) Thing Description [[wot-thing-description11]].
+
[Term]
[To be added]
@@ -251,7 +282,7 @@
Visions of Agents on the Web
Recent years have brought renewed interest in Web-based MAS — as evidenced by the Dagstuhl Seminar 21072 (Feb. 2021) and Dagstuhl Seminar 23081 (Feb. 2023) on "Agents on the Web", which led to the creation of the W3C Autonomous Agents on the Web (WebAgents) Community Group. A key enabler for this renewed interest is the Web of Things, which provides new practical use cases for Web agents and realizes several visionary ideas anticipated in the original Semantic Web paper [[SEMWEB01]]. Another key enabler is the recent progress in LLM-based agents that can follow instructions and use tools: just like previous generations of agents, LLM-based agents are designed for specific tasks, underscoring the need for open networks in which agents complement one another's abilities to solve more complex problems. New protocols and frameworks are emerging to support LLM-based agents to discover and use tools, or to discover and interact with other agents — many of them explicitly building on Web standards to foster interoperability (e.g., see the Model Context Protocol, Agent2Agent Protocol, Agent Network Protocol, Eclipse LMOS).
-
Conceptual Dimensions for Web-based Multi-Agent Systems
+
Conceptual Dimensions
A multi-agent system (MAS) has several distinguishing features. One key feature is decentralized control, where each agent makes its own decisions and controls its own behavior — yet the MAS as a whole exhibits coordinated behavior to achieve system-level design objectives. Another key feature is that capabilities, knowledge, and resources are distributed among agents, which creates inter-dependencies: agents participate in a MAS because they need to interact with one another to solve problems that would otherwise exceed their individual capacities. Without such inter-dependencies, the MAS would be a collection of isolated agents — and would not constitute a system at all.
@@ -281,6 +312,39 @@
Conceptual Dimensions for Web-based Multi-Agent Systems
Throughout this report, we use these four conceptual dimensions to organize the discussion and emerging technologies.
+
+
Architectural Considerations
+
+
+
+
+
Design Goals
+
+
+
+
+
+
+
Architectural Patterns
+
+
+
+
+
+
+
Considerations
+
+
+
+
State of Web-based Multi-Agent Systems
@@ -435,13 +499,6 @@
Agentic AI
-
-
Architectural Considerations
-
-
-
Identification
@@ -589,7 +646,7 @@
Discussion
-
Norms, Policies, and Organizations
+
Policies, Norms, and Accountability
-
Conclusions: A Strategy for Agents on the Web
-
+
Conclusions: A Roadmap for Agents on the Web
+
+
+
Accountability
+
+
Make a case for accountability; what do we need to enable accountability, e.g. transparency? answerability (building a dialogue)?