For AI agents: a documentation index is available at the root level at /llms.txt and /llms-full.txt. Append /llms.txt to any URL for a page-level index, or .md for the markdown version of any page.
Contact SalesGet Support
HomeGuidesAPI reference
HomeGuidesAPI reference
  • Get started
    • Quickstart—Verify a Business
    • Quickstart—Register an entity for payroll taxes
    • Learn how Middesk works
    • Security
    • Changelog
  • Build with Middesk
    • Get your API keys
    • Understand API changes
    • Status codes and errors reference
    • Implement webhooks
    • Secure webhooks
    • Use GraphQL
    • Connect MCP
  • Verify a business (KYB)
    • Verify TIN
    • Verify name and address
    • Verify Secretary of State status
    • Verify owners and officers
    • Screen for sanctions and watchlists
    • Search for adverse media
    • Search for Politically Exposed Persons
    • Implement KYC
    • Discover connections
    • Evaluate online presence
    • Assess credit risk
    • Accelerate onboarding
    • Monitor business activity
    • Manage business entities
    • Work with agents
      • Run agents
      • Stream run events
      • Take action
      • Interrupt policy
      • Webhook events
LogoLogo
Contact SalesGet Support
On this page
  • Interacting with agents
  • Responding to agents
Work with agents

Take action

Was this page helpful?
Previous

Interrupt policy

Next
Built with

Agents are designed to work alongside you, not replace your judgment. When an agent completes its analysis, it hands off to you with structured findings and recommendations. From there, you review the results, provide input, and direct the agent to carry out the next steps. This interaction model keeps you in control of your decisions while reducing the manual effort to reach them.

Interacting with agents

Depending on the use case, you can require an agent to pause and wait for input using a Human-in-the-loop (HITL) pattern. To support this, agents support interrupts.

An interrupt is a step with type interrupt. It signals that the agent has paused execution and is waiting for input before continuing. The interrupt step includes the agent’s findings as artifacts with confidence scores and source references.

Responding to agents

Agent Action Flow

A user runs an agent which plans and executes. The agent interrupts with sources to add, the user approves, then the agent adds sources, refreshes the report, and approves the business.

UserAgentClientRun an AgentCreates agent runAgentPlanAgentExecuteInterrupts withsources to addClientApproves sourcesResumes agent runAgentAdds sourcesAgentRefreshes reportOutcomeApproved

In responding to an interrupt, you delegate an action to the agent. Actions represent the decisions and modifications you make to a Middesk report. Available action types include:

TypeDescription
decisionApprove or reject the business
dismissalDismiss a watchlist result as a false positive
verify_tinMark a TIN as verified via document
add_sourcesAdd external data sources to the business
add_attributesAdd addresses or people to the business

For the full action schema and type-specific payloads, see the Action reference.

Get a demo
Contact your account manager or contact sales to inquire about access.