Build Autonomous AI Agents, Native to Your Supply Chain

Salesforce Agentforce and Microsoft Copilot Studio are broad agent-building platforms designed to support many business functions. Supply-chain platforms such as Blue Yonder also provide pre-built agents for specific operational use cases.

Agent Foundry is purpose-built for supply chain and commerce. It gives teams a no-code starting point, with deeper technical control when needed, to build, modify, and deploy agents around their own workflows. Those agents operate on the unified ActivePlatform foundation, using Manhattan’s operational data, business services, and approved action surfaces to move from context to governed action.

The result is a practical way to extend the Manhattan Active platform with customer-specific workflows, without asking teams to recreate their supply-chain context in a separate AI environment.

No. Getting started doesn’t require an AI engineer or prompt engineering. You can start by describing the job in plain language or adapting a template. Developers and technical teams can take deeper control of logic, tools, orchestration, and integrations when the use case requires it.

No. General-purpose generative AI tools and large language models (LLMs) can provide the language model and generate an answer, but by themselves they do not know the live state of your operation or which actions your business allows. Agent Foundry uses those models within the Manhattan Active platform, grounding agents in Manhattan’s operational data and services and connecting them to approved workflows and APIs. That lets an agent reason about a supply chain or commerce situation and take a bounded, auditable action, rather than stopping at a response.

LangChain, Claude Code, and Claude AI can be powerful tools for developers, but they leave you to assemble the operational context, data access, tool calls, workflow logic, permissions, approvals, and lifecycle management around the model. Agent Foundry provides a Manhattan-native path for building agents: start with natural language or a template, ground the agent in ActivePlatform data and services, connect it to approved actions, and manage it through build, test, deployment, and ongoing change. Developers can still take deeper control when needed, but the business does not have to recreate Manhattan’s supply chain and commerce context outside the platform.

No. You do not need to train or operate your own AI model or manage machine learning to build agents with Agent Foundry. The platform combines LLM reasoning with Manhattan business logic, workflows, APIs, and tools. You configure what the agent sees, how it reasons, what it is allowed to do, and where approval or human review is required. Advanced users can apply prompt engineering, select data sources, and define multi-step workflows when a use case needs deeper control.

Google Antigravity is a developer-oriented platform for coding agents. Agent Foundry serves a different purpose: helping supply chain and commerce teams build, configure, and govern agents around operational workflows.

With Agent Foundry’s dual-path agent development approach, both business practitioners and professional developers can create new agents.

For business practitioners without coding knowledge, the platform offers a build-by-template method. Users can express their instructions in plain English, such as asking an agent to check load utilization during planning. The technology will then automatically suggest relevant pre-built “helper agents” from its component library to facilitate the creation of the new agent.

Advanced users and developers have the option to take it further by building agents from scratch. Utilizing the advanced development features of Manhattan Agent Foundry, developers can undertake more complex tasks. This includes fine-tuning reasoning prompts for specialized scenarios, selecting Manhattan Active APIs or other data sources, and defining multi-step workflows.

This dual-path approach provides both business-friendly simplicity and developer-level control, allowing our customers to innovate at their own pace.

Agent Foundry combines flexible AI reasoning with controlled execution. Guardrails, approved service definitions, conditional workflow logic, permissions, content-sanitization controls, and human approvals help ensure agents access only authorized data and take only approved actions. This gives teams a governed path to scale agentic workflows with confidence.