Skip to main content
Feedback

Agent Garden REST APIs

The Boomi Agent Garden REST API lets you programmatically create, configure, package, and deploy AI agents within your Boomi account. Use it to automate the full agent lifecycle, from creation to live deployment in a runtime environment, without touching the UI.

The API follows REST conventions and uses JSON for all request and response payloads. Two API groups cover the full lifecycle:

The Agent Garden API supports the following core operations:

  • Agent management: Using the API you can create agents by specifying objectives, tasks, guardrails, and custom configurations. You can also retrieve paginated lists of agents with optional filtering or fetch the detailed configuration of a specific agent.
  • Packaging and deployment: You can generate snapshots of an agent's configuration and prepare it for deployment into an environment. A package encapsulates the complete agent definition at the time of creation. You may optionally trigger immediate deployment to an environment during the packaging request. Packages can be listed per agent or account-wide and are eligible for deletion provided no active deployments reference them. Once the agent is packaged you can deploy it to a target environment and make it available to others. Deployment records are retrievable per agent, per package, or across the account. Specific deployments are removable when they are no longer required.
  • Agent execution: Execute APIs handle the runtime layer of the Agent Garden invoking agent sessions.

Use case : Multi-Unit Enterprise Agent Deployment

A large financial services firm has 12 regional branches plus a central support team. After building and testing a support-triage agent in Agent Studio, they need to roll it out company-wide. The agent handles tier-1 ticket routing, policy lookups, and escalation recommendations.

But there’s a catch: different teams have different constraints.

  • HQ support team has deep integration needs (CRM connectors, custom-logging) and engineering resources.
  • Regional branches lack engineering expertise and need something fast. They want to embed an agent widget on their websites with minimal setup.
  • IT/DevOps wants all deployments managed through their CI/CD pipeline, with version control and automated rollouts across environments.

The Agent Garden API lets them do all three at once - same agent with three deployment paths.

Choose Your Deployment Path

Path 1: Execute API (Flexible & Custom)

Best for: Teams that need full control over the chat UI, custom integrations, or advanced analytics.

Gives you the ability to:

  • Direct control over session lifecycle
  • Build your own chat interface (React, Vue, vanilla JS, native mobile, etc.)
  • Custom-logging, analytics, and CRM integration
  • SSE streaming or JSON polling for real-time responses

Example: HQ support team embeds the agent inside their proprietary CRM platform. They thread agent responses into their case-management system and log interactions in a custom data warehouse.

When to pick this path:

  • You need a custom chat UI that matches your brand or product design
  • You are integrating with proprietary systems (CRM, ticketing, data warehouse, etc.)
  • You need custom analytics or audit logging
  • You want to control the exact user experience

Next step: Execute API documentation

Path 2: Design API + CI/CD Pipeline (Automated & Scalable)

Best for: Organizations managing multiple deployments across dev, staging, and production that require version control and automated rollouts.

Gives you the ability to:

  • Achieve programmatic agent creation and configuration
  • Package agents as versioned snapshots
  • Automate deployment to multiple environments
  • Conduct a full audit trail and rollback capability
  • Not UI clicks abd handle everything in code

Example: DevOps maintains agent definitions in a Git repository. On merge to main, the CI/CD pipeline automatically packages the agent and deploys it to staging, runs tests, then promotes to production across all 12 regional environments.

When to pick this path:

  • You manage multiple deployments (dev, staging, production)
  • Your team uses Git for version control and CI/CD for releases
  • You need consistency, immutability, and rollback capability across environments
  • You want to treat agent configuration like code

Next step: Design API documentation

Path 3: EmbedKit CDN (Simple and Quick)

Best for: Teams without engineering resources. Rapid deployment. No custom UI needed.

Gives you the ability to:

  • Pre-build with a themeable chat widget
  • Have one-line embed code
  • Work on any website by being CORS compatible
  • Not require build tools, npm packages, or API integration

Example: Regional branch offices embed the agent on their public-facing website with a single JavaScript snippet. No custom integration needed.

When to pick this path:

  • You need to launch in days, not weeks
  • Your team doesn't have frontend engineering capacity
  • A standard chat widget meets your UX needs
  • You are comfortable with Boomi's built-in theming and customization options

Next step: EmbedKit CDN documentation.

The Hybrid Approach

Many organizations use both EmbedKit and Execute API depending on the use case:

  • EmbedKit for public-facing, self-service experiences (branch websites, partner portals)
  • Execute API for internal tools and custom integrations (CRM, support platforms, internal dashboards)
  • Design API managing the agent across both, ensuring consistency

Next Steps

  1. Unsure which path fits your needs? Start with Agent Garden API overview.

  2. Ready to embed quickly? Refer to EmbedKit CDN documentation.

  3. Building a custom integration? Refer to Execute API reference.

  4. Managing deployments at scale? Refer to Design API reference.

On this Page