Deterministic workflows for Claude Code

Convert daily tasks into workflows you can trust, repeat, and share.

flowwork turns your daily work into auditable, repeatable programs that drive Claude Code — combining deterministic orchestration with the generative power of agents in one durable unit.

Free tier: 3 active workflows · 10 runs/day · no credit card

See flowwork in action

Short, focused demos of real workflows driving Claude Code.

A quick tour: author a workflow in Claude Code, deploy it with flowctl, and watch a run execute end to end.

Why flowwork

The flexibility of a coding agent with the guarantees of a versioned, strongly-typed workflow.

No MCP sprawl

External calls execute inside the workflow. You pay the token cost once, at authoring time — not on every agent run.

MCP import docs →

Stay in your Claude Code session

Keep working in your Claude Code session while getting the predictability of versioned, strongly-typed workflows.

Full power of Claude Code

Skills and built-in tools are callable from a workflow. Drive Claude Code over local files, docs, devops, source code, and browsers — on your laptop or in an ephemeral sandbox you spin up, run, and tear down for cents.

Reusable & auditable

Workflows are repeatable, shareable units with a full audit trail. Run the same thing twice and get the same shape of result.

Resilient connectors

Retries, exponential backoff, and greedy fallback out of the box — or author your own graceful degradation.

Build on top

Develop custom applications and agents on top of flowwork. Integrate your own app as an MCP server or CLI tool — the engine is the durable substrate; you bring the product.

MCP import docs →
Security, by construction

The strongest case for flowwork is what you don't have to defend.

Because the code is prewritten, there is no sandboxing to bolt on, no AI-firewall trying to out-guess a probabilistic model, and no class of prompt-injection that turns into an arbitrary external call. Those defenses are inherently insufficient against the nondeterminism of LLMs.

Every external call is abstracted into a connector with explicit checking, retries and backoff, and a fallback you control. The model decides what to do within bounds you authored — it never decides how to reach the outside world.

How it works

01

Author

Describe the workflow in words and use the 'flowwork' skill to generate a typed, versioned definition — connectors and all.

02

Operate

Trigger workflows from Claude, monitor execution, and handle human-in-the-loop steps as they come up.

03

Audit

Execute it repeatably. Every run is auditable, resumable, and shareable as one unit.

Use cases

DevOps automationDocument operationsSource-code automationBrowser automationInternal Tools and AutomationsCustom Workflow-centric Applications

Simple pricing

Start free. Own it with a one-time Pro license. Scale on Cloud.

Free

$0

For trying it out

  • ✓ Up to 3 active workflows
  • ✓ 10 runs per day
  • ✓ Full engine, runs locally
  • ✓ Built-in connectors · Basic skill
Start free
Early adopter · one-time

Pro

$49 one-time

Early-adopter price — locks in for life; rises as features land.

  • ✓ Unlimited workflows & runs
  • ✓ Pro authoring skill
  • ✓ MCP import — any server, unlimited tools
  • ✓ Workflow Garden
  • ✓ Everything in Free + a year of updates
Get started
Coming soon

Team / Cloud

Recurring

For teams and heavy, multi-machine workloads

  • ✓ Everything in Pro
  • ✓ Hosted inference server
  • ✓ Workspaces — shared state & memory
  • ✓ Workflows across multiple machines
  • ✓ Multi-seat & team management
Coming soon
Capability Free Pro Team / Cloud
Active workflows 3 Unlimited Unlimited
Runs per day 10 Unlimited Unlimited
Authoring skill Basic Pro Pro
MCP import — any stdio / remote server, unlimited tools per workflow
Workflow Garden — search & import community workflows Soon Search
Machines per license (run on laptops, servers & ephemeral sandboxes) 1 5 Unlimited
Workspaces — shared state & memory across machines Soon
Inference server — hosted Soon

Items marked Soon are on the roadmap and ship to the relevant tier as they land. Team / Cloud pricing announced at launch.

Questions

What do the trial limits mean? +

The free tier allows up to 3 active workflows and 10 runs per day — enough to build something real. Upgrade to remove the limits.

Why is this more secure than an agent with a bunch of MCPs? +

Because the code is prewritten. There's nothing to sandbox at runtime and no AI-firewall guessing at a probabilistic model's intent. External calls are abstracted into vetted connectors with explicit checking.

Why not just use a database for durability? +

You can — until you need retries, backoff, leases, versioning, fan-out, long timers, replay, and operator tooling. Then you've rebuilt a workflow engine. We wrote about this.

Does it work offline? +

Yes. The engine runs locally against your own Postgres. The trial license checks in periodically but tolerates being offline within a grace window.

Build your first workflow today

Browse the community workflow collection, watch a demo, then start your free trial.