Runtype
For Engineers

Stop building the platform.
Start building the product.

Runtype gives you agents, flows, multi-surface deployment, and a security model — so you build product logic, not orchestration infrastructure.

The Problem

The model call is one line. Everything else is the job.

You can call an LLM in five minutes. Then you spend the next month building orchestration, credential management, tool execution, surface adapters, state persistence, and deployment infrastructure. That’s not product work — that’s platform work.

Orchestration and step sequencing

Credential isolation from the model

Tool execution across environments

Adapters for every deployment channel

Execution state and resumability

Cost tracking and rate limiting

The Platform

The platform layer, already built.

One product, many surfaces

Define your agent or flow once. Deploy it to web chat, Slack, API, email, MCP, webhooks, A2A, and more. Each surface gets its own auth, routing, and configuration. Adding a surface is a config change, not a rewrite.

Protected parameters

Credentials are hidden from the model context and injected at execution time. The agent calls your API without ever seeing the secret. No wrapper functions, no manual scrubbing — it’s the default behavior.

Hybrid tool execution

Agent reasoning runs in the cloud. Tools execute wherever your data lives — browser, local machine, or on-prem server. The runtime handles the pause and resume across environments. Sensitive data never has to leave its origin.

Runtime tool composition

Define which tools an agent can use at dispatch time, based on user, tenant, or environment context. Different customers get different capabilities from the same agent — without forking the logic.

Agent loops with cost controls

Multi-turn reasoning with configurable iteration limits, cost budgets, and reflection intervals. Agents observe, think, act, and observe again — with guardrails that prevent runaway execution.

Managed edge runtime

Runs on Cloudflare Workers globally. Zero cold starts. Durable Objects for stateful coordination. You get the infrastructure without managing the infrastructure.

Architecture

ProductCapabilitySurface

Runtype’s core abstraction is the product. A product composes capabilities (agents and flows) and exposes them through surfaces (deployment channels). Define the product once. The runtime handles execution, routing, auth, and delivery per surface.

1

Define product logic

Agents for reasoning, flows for orchestration, tools for action.

2

Compose capabilities

Each capability is a named operation backed by an agent or flow.

3

Attach surfaces

Chat, API, Slack, email, MCP, webhook, A2A — each independently configured.

4

Ship with runtime controls

Protected params, cost budgets, rate limits, execution logging.

The Difference

A runtime, not another library.

Frameworks give you building blocks and leave you to host, operate, and scale them. Runtype is a managed runtime — execution, state, security, and deployment are handled. You write product logic. The platform runs it.

Framework
Runtype
You host and operate
Managed runtime
You build surface adapters
Multiple surfaces built in
You handle credential isolation
Protected parameters by default
You wire up tool execution per environment
Hybrid execution built in
You implement cost controls
Cost budgets and rate limits included
You stitch together observability
Execution logging and live tail
Examples

Ship real features, not side projects.

Authenticated internal agent

Agents that act on internal systems — CRM, ticketing, databases — without exposing credentials to the model. Protected params handle the secrets. Runtime tools scope the capabilities per user.

Multi-channel support system

One agent handles support across web chat, Slack, and email. Same logic, same tools, same state — different surfaces with surface-appropriate formatting.

Scheduled batch processor

Agents that run on a schedule, process records in batch, and write results back. Cost budgets prevent runaway execution. Execution logs give you an audit trail.

FAQ

Common questions

Can I use my own models and keys?

Yes. Multi-provider routing across OpenAI, Anthropic, Google, xAI, open source, and more. BYOK or use platform keys to start immediately.

Is this just a wrapper around model APIs?

No. The model call is one step. The value is everything around it — orchestration, security, tool execution, multi-surface deployment, state management, and operational controls.

Can I control how tools and flows behave?

Full control. Runtime tool composition lets you define capabilities per dispatch. Flow steps support conditional branching, error handling, and custom transforms.

Does it work in my stack?

The API is HTTP + SSE. SDKs in TypeScript. MCP server for Claude and IDE integration. CLI for local development. Embed via script tag or React component.

Get Started

Build the product layer, not the glue layer.