Skip to content

OpenAI Agent Builder alternative: workflows that run as code

Agent Builder is OpenAI’s visual canvas for multi-step agent workflows, and OpenAI has scheduled it to shut down on November 30, 2026. Whenever writes the workflow as code from your description or your coding agent, and calls a model only in the steps that need one.

By Whenever team · Updated · Workflow behavior and runtime limits

What you get on Whenever

  • Code instead of a canvas

    Describe the workflow, or ask Claude Code, Codex or Cursor to write it. What you review is a TypeScript workflow module.

  • A model only where a step asks

    Fixed steps run as code without a model call; a step that needs judgment can still call one.

  • Hosted and triggered for you

    Schedules, webhooks, app events and manual runs, with credentials you enter once in Setup.

Model usage, or a run per workflow.

OpenAI includes AgentKit tools with standard API model pricing. Whenever counts whole workflow runs, with optional runtime AI on top.

OpenAI Agent Builder

Standard API model pricing

Model usage

OpenAI says AgentKit tools are included with standard API model pricing; Agent Builder is scheduled to shut down on November 30, 2026.

Whenever

$39/mo

10,000 runs per month

Find the right plan for you

Tax exempt for this offering.

At full use of 10,000 included runs, $39 ÷ 10,000 ≈ $0.0039/included run. This is an allowance average, not a per-run tariff, overage price or competitor savings claim. The monthly subscription is still $39 if you use fewer runs; unused allowance, runtime AI usage, moving to a larger plan and external charges change your effective cost.

OpenAI includes AgentKit tools with standard API model pricing. Compare that with Whenever runs and optional runtime AI; no competitor dollar price is asserted.

A run is one execution of one workflow, start to finish. AI building and editing are free within weekly usage limits; runs are metered separately.

A closer look

What changes when you describe the job in plain language and code runs it: to build, to run, and to keep.

Repeatable execution logic

OpenAI Agent Builder

OpenAI describes an Agent Builder workflow as a combination of agents, tools and control-flow logic, composed from connected nodes.

Whenever

AI writes the procedure; published code runs its fixed logic. Optional AI steps can vary, and external data, time or randomness can change results. Fixed logic does not guarantee exactly-once delivery.

Runs per dollar

OpenAI Agent Builder

OpenAI says AgentKit tools are included with standard API model pricing.

Whenever

Pro includes 10,000 runs per month; see the pricing page for current US dollar prices. AI building and editing are free within weekly usage limits. Published runs are metered separately: the meter counts the whole run, not individual steps. External services and optional model calls can have their own charges.

What you can express

OpenAI Agent Builder

A visual canvas with templates, drag-and-drop nodes with typed inputs and outputs, tools and guardrails.

Whenever

Describe the rules you already know; AI writes the procedure. Review the summary and sample results, not necessarily the code. Keep an explicit model step for judgment. Agent platforms may also offer fixed workflows or scripts; compare the mode you actually use.

Export and portability

OpenAI Agent Builder

A published workflow is a versioned object with an ID for ChatKit, or you can download the Agents SDK code to run it yourself.

Whenever

The TypeScript is yours: read it, edit it, export it, take it with you. Moving it to another runtime can require adapting managed connections, triggers and runtime-specific integrations.

Debugging a failed run

OpenAI Agent Builder

Preview runs your workflow interactively and shows the execution of each node; trace graders assess the workflow.

Whenever

Each step’s recorded result, a debugger with breakpoints on a saved version, and an email on the first failure. A failed run is not retried or resumed for you.

Where the AI cost lands

OpenAI Agent Builder

Usage is billed at standard API model pricing, as OpenAI states for AgentKit tools.

Whenever

AI writes the workflow once, with free building subject to weekly AI usage limits. A published procedure runs its steps without a model choosing them. Any model call you explicitly include still runs and has its own cost.

Availability

OpenAI Agent Builder

OpenAI is deprecating Agent Builder; it is scheduled to shut down on November 30, 2026. ChatKit remains available.

Whenever

Whenever does not import Agent Builder workflows. Describe the workflow or ask your coding agent to rewrite it, then test it before switching.

Monitoring and notification workflows to start from

Starter briefs for related jobs, not recorded runs. Connect authorised accounts, confirm trigger and plan requirements, then review and test the workflow before publishing.

Browse all templates

Questions people ask before switching

Is OpenAI Agent Builder shutting down?

OpenAI has announced that Agent Builder is deprecated and scheduled to shut down on November 30, 2026. ChatKit remains available.

Sources and scope

Official sources captured on the date shown below. These describe published features and billing rules, not independent performance tests. Check the linked documentation for current modes, plans and limits. No competitor dollar price is asserted.

Checked

Move one Agent Builder workflow

Describe it, or ask your coding agent to rewrite it. Whenever hosts, triggers and runs the code you reviewed.

  1. Record what the workflow does

    Its inputs, each node’s job, the tools it calls and where its output goes.

  2. Rewrite it as a workflow

    Describe it in plain words, or ask your coding agent to follow www.whenever.dev/skill.md. Keep a model step only where the work needs judgment.

  3. Test, then publish

    Enter the credentials in Setup, run it once by hand, then publish on its trigger.

Start for freeNo credit card requiredAI building has weekly usage limits.