Interoperabilidade de agente

Interoperabilidade de agente enables Automation Anywhere to connect with third-party AI assistants and agents using the Model Context Protocol (MCP). It allows Agentes de IA to discover and invoke automations, processes, and goal-based Agentes de IA, securely across systems.

Benefits

You can connect Automation Anywhere Agentes de IA with agents from other companies. This allows for smooth teamwork across different platforms. It helps avoid isolated systems, keeps the system up-to-date, and ensures it can grow easily—without being tied to one vendor.

  • Seamless collaboration across Automation Anywhere Agentes de IA, third-party Agentes de IA, and enterprise systems.
  • Lower integration costs by eliminating custom point-to-point connectors.
  • Dynamic discovery of tools and actions at runtime.
  • Reliable automation with built-in context continuity across calls.
  • Unified governance and security through centralized authentication, logging, and access control.
  • Composable orchestration for flexible, multi-agent workflows.

Model Context Protocol (MCP)

Model Context Protocol (MCP) is an open protocol that standardizes how Agentes de IA:
  • Discover tools
  • Understand capabilities
  • Execute actions across systems

Instead of building custom, point-to-point integrations, MCP provides a consistent contract for agents to request context, invoke tools, and receive structured results.

MCP overview

Note: The Microsoft, Salesforce, ServiceNow, SAP, and MCP marks and logos are trademarks or registered trademarks of each company, respectively, and are used for identification purposes only.

MCP defines how tools are described, how inputs and outputs are exchanged, and how execution is governed. This allows Agentes de IA, both Automation Anywhere Agentes de IA and third-party, to reliably interact with automations, APIs, and enterprise workflows without tight coupling.

For enterprises, MCP acts as a control plane for agent interactions. It enables secure access, policy enforcement, observability, and scalability as organizations move from isolated AI Assistants to coordinated, multi-agent systems. In practice, MCP turns AI Agents from passive copilots into actionable participants in real business processes.

What is an MCP Inbound Tool / Agent connection?

An MCP inbound tool, which is configured as an Agent connection, allows third-party AI Assistants and AI Agents to communicate with and execute Automation Anywhere automations. By standardizing how AI Agents discover, understand, and execute actions, MCP facilitates seamless collaboration across different platforms and vendors. This reduces integration costs and provides dynamic discovery of tools at runtime. Automation Anywhere positions itself as the orchestration and execution layer for complex, high-agency work.

Our MCP inbound interactions are powered by our Process Reasoning Engine (PRE). See Mecanismo de raciocínio de processos e IA generativa. The PRE has a deep understanding of all the Automation Anywhere automations, processes, API Tasks, and Agentes de IA. It knows what they are and what they can do. When a user requests to use an MCP inbound tool from a third-party AI Assistant, our PRE understands the user's intent and the context of the request. Then, it matches this with the unique automation in the Automation Anywhere repository.

With inbound connectivity, Automation Anywhere is receiving a request from third-party Agentes de IA and third-party AI Assistants.

Availability

The following Agent Interoperability capabilities are available based on your license type:

For Cloud customers:
  • The Base platform license provides core Agent Interoperability. This allows third-party AI assistants to trigger Automation Anywhere automations using MCP inbound. It helps customers connect external Agentes de IA to existing automations with minimal setup.
  • The Enterprise platform license includes everything in the Base license, plus advanced interoperability features, like PRE integration. PRE automatically creates descriptions for automations and their input and output variables that both humans and agents can read. This makes it easier to find MCP tools and helps third-party Agentes de IA understand what an automation does, what inputs it needs, and what outputs it returns. This leads to more accurate and reliable agent interactions.
  • Enterprise customers also get access to predefined MCP tools, such as DiscoverAutomation. These tools allow authorized users from an MCP client to discover and access automations in the Automation Anywhere repository without manually creating agent connections, simplifying onboarding and accelerating agent-to-automation adoption.

Traditional APIs versus MCP Inbound

Feature Traditional APIs MCP Inbound
Tool discovery Static, predefined; require agents to know endpoints and parameters upfront Dynamic, runtime discovery; agents discover capabilities at runtime
Context Manual management Automatic continuity; maintains context continuity and governance across calls
Orchestration External, brittle; APIs break easily with unexpected inputs Built-in, composable
Vendor support Locked-in Cross-vendor, open
Security Fragmented Centralized governance