Skip to content

For founders & builders

Ship the automation. Skip the job servers.

Signed webhooks, HTTP requests, cron schedules, sub-workflows and error handlers on a visual canvas. Run AI on any of 14 providers with your own key, and test whole flows with Dry Run before anything real is sent.

Start free See pricing

What changes

Less copying, chasing and checking.

  • Today

    Every small automation needs a cron job, a queue worker and somewhere to host them.

    With Fluxo Flow

    Hosted schedules (cron included) and webhook triggers, with run history for every run.

  • Today

    Your AI costs are tied to one vendor and someone else's markup.

    With Fluxo Flow

    Bring your own key for any of 14 providers, including any OpenAI-compatible endpoint, and pick the model per step.

  • Today

    Testing an automation means hitting production APIs.

    With Fluxo Flow

    Dry Run walks the whole graph in your browser with side effects and AI calls simulated.

  • Today

    Failures disappear into logs nobody reads.

    With Fluxo Flow

    Retries with backoff, Error Handler steps, step-by-step traces and run comparison.

Example workflows

Three workflows you could build this week.

Each one uses nodes that exist on the canvas today. Start from a blank canvas or a marketplace agent and adapt it.

Signup enrichment

5 steps

Your backend posts a signed event when someone signs up. The workflow enriches it and routes good-fit accounts.

  1. Step 1: Webhook triggeruser.signed_up, HMAC-signed
  2. Step 2: HTTP RequestCall your enrichment API
  3. Step 3: AgentScore fit against your ideal customer
  4. Step 4: RouterGood fit?
  5. Step 5: Slack PostPing the founder channel

Result: You hear about the signups worth a personal email, while they're still fresh.

Nightly sync with recovery

5 steps

A cron schedule pulls records, processes each one in a shared sub-workflow, and reports failures.

  1. Step 1: Scheduled triggerCron 0 2 * * *
  2. Step 2: HTTP RequestFetch changed records
  3. Step 3: LoopFor each record
  4. Step 4: Sub-workflowUpsert the record
  5. Step 5: Error HandlerPost failures to Slack

Result: A sync you can trust, with failures reported instead of hidden.

Answers grounded in your own data

5 steps

A webhook asks a question. The workflow looks up relevant notes, answers, and checks the shape of the result.

  1. Step 1: Webhook triggerQuestion from your app
  2. Step 2: RetrieverFind relevant entries
  3. Step 3: AgentAnswer from those entries only
  4. Step 4: ValidatorCheck against a JSON schema
  5. Step 5: Publish OutputReturn to your webhook

Result: Structured answers your app can rely on.

Details

The details builders ask about.

  • Webhook security

    Shared secret header, HMAC-SHA256 signature, or both. Optional IP allowlist and required headers.

  • Real limits, stated

    Each server run has a time limit of about 50 seconds and stops if its estimated AI cost passes $2.

  • Versions and export

    Every save records a version you can restore, and any workflow exports as a file you can import elsewhere.

Features

The parts you'll use most.

Plans

Which plan usually fits.

Explore free with no credit card and Dry Run your workflows. Upgrade to run them live, with more workflows, seats and faster schedules.

  • Usually the best fit

    Basic$39/mo

    A handful of production workflows running live, on schedules, webhooks or the API.

    • 15 workflows
    • 3 seats
    • Schedules as often as every 15 minutes

  • Also a good fit

    Pro$119/mo

    When you need more workflows, faster schedules, or teammates on the canvas.

    • 25 workflows
    • 5 seats
    • Schedules as often as every 5 minutes
    • 2 marketplace agents included

Explore on Free with no card: build up to 3 draft workflows and Dry Run them, then upgrade to run them live. Compare every plan

FAQ

Questions, answered.

Is there a public API?

Yes. A REST API with scoped API keys starts runs (with idempotency keys for safe retries), returns step-by-step results and cancels runs, and signed outbound webhooks tell your systems when runs succeed, fail or wait for review. It's included on every paid plan, with per-plan rate limits. You can also start runs by POSTing JSON to a workflow's webhook URL.

Can I use a self-hosted or niche model?

Yes, through a custom OpenAI-compatible endpoint, as long as it's reachable at a public https address. Local and private network addresses are blocked.

How do I avoid processing the same event twice?

Record each event ID in a persistent Memory node, then use a Router at the start of the next run to skip IDs you've already handled.

What are the run limits?

Each server run has a time limit of about 50 seconds and a $2 cap on estimated AI cost. Plans limit the number of workflows, seats and how often schedules can run.

Ship your first workflow tonight.

Build and Dry Run on the Free plan, no credit card. Run live on a paid plan, on your own AI key at cost.

Start free See pricing

Free to explore · No credit card · 30-day money-back guarantee on paid plans