Skip to content
Product

One platform for the work between systems.

Build it on a canvas, run it on your servers, publish it any way your people need it.

Visual pipelines

A pipeline is a flow chart that actually runs. Put each step on a canvas, connect them, and independent steps run at the same time — no glue code, no scripts on someone’s laptop.

Explore visual pipelines
  • Conditions, validation rules and row-level rejects
  • Parallel branches, for-each loops and merges
  • Per-step retries, timeouts and failure paths
  • Draft / published versions with rollback
  • Run history drawn on the canvas, step by step

Connect everything

SpanFlow talks to databases, HTTP APIs, file stores, mailboxes, directories and servers out of the box — and any web application you can describe in a simple manifest.

Explore connect everything
  • PostgreSQL, SQL Server, MySQL, SQLite
  • HTTP/REST with OAuth2 and auto-pagination
  • Local, FTP, SFTP, Amazon S3, Azure Blob storage
  • Email via SMTP or SendGrid, with attachments
  • LDAP / Active Directory

Triggers & schedules

Start a pipeline on a schedule, when another system calls a URL, when something new appears, or from another pipeline. One incoming URL can route events to many pipelines.

Explore triggers & schedules
  • Cron schedules with time-zone support
  • Inbound webhooks with a smart router
  • Sync jobs: poll → filter → deliver, with incremental state
  • Sequences: run several pipelines as one ordered unit
  • Run queue with workers and concurrency limits

Publish as an API

Publish a pipeline as a real HTTP endpoint. Group endpoints into an App with its own URL space, credentials and documentation — then put it on your own domain.

Explore publish as an api
  • Published endpoints with input validation and file uploads
  • API keys, token endpoint, IP allowlists, rate limits, CORS
  • Auto-generated OpenAPI docs and call snippets
  • Apps: one URL space and credential set for many endpoints
  • Custom domains and platform subdomains

Front ends included

Give every published app a user interface. Scaffold a form from an endpoint, design screens on a drag-and-drop canvas, describe a page to AI, or upload your own React build.

Explore front ends included
  • Form scaffolding from an endpoint’s inputs
  • Drag-and-drop UI designer with page state and actions
  • AI page authoring with a one-shot checker
  • Upload a React / Vite build as the app’s front end
  • Shareable preview URLs and per-environment deploys

AI built in

Use language models as ordinary pipeline steps: summarise, classify, extract fields from PDFs and images, let an agent choose tools, or search your own data. Bring your own model provider.

Explore ai built in
  • LLM step with JSON mode and templated prompts
  • Extract: text and structured fields from PDFs and images
  • Agent step that picks tools within set limits
  • Embeddings and vector search (pgvector)
  • MCP client and MCP server / data gateway

Data mapping

Map fields from one shape to another visually — one company’s invoice into another’s, a legacy export into a modern API. Save mappings once and reuse them in any pipeline.

Explore data mapping
  • Visual mapping designer
  • Works on any records a pipeline reads — CSV, Excel, JSON, database rows, API responses
  • Joins, grouping, aggregates, conditions and dozens of functions
  • Versioned, shareable mappings
  • Value maps (lookup tables) for code translation

Chat flows & campaigns

Design chat flows on a canvas that call your pipelines — check an order, book a slot, look up a balance — then run them on your website, WhatsApp or SMS, and send campaigns to the people who opted in.

Explore chat flows & campaigns
  • Visual flow designer with a sandbox tester
  • Environments and deploys per flow
  • WhatsApp (Meta Cloud API), SMS and HTTP channels
  • Campaigns, recipients and analytics exports
  • Client SDKs for web and .NET

Monitoring & alerts

Every run is recorded step by step. Watch activity live, drill into any payload, and push alerts to the people responsible — on email or straight to a phone.

Explore monitoring & alerts
  • Per-step run history with payloads
  • Live activity and observability views
  • Usage analytics for every API call
  • Email notifications and alert inbox
  • Device push to a companion app

Security & self-hosting

SpanFlow is self-hosted: it runs inside your network, next to your systems. Organisations, projects, roles and permissions keep every team’s work separate and every action accountable.

Explore security & self-hosting
  • Docker deployment with health checks
  • Multiple instances and worker pools
  • Organisations, projects, roles and permissions
  • Google, GitHub and LinkedIn sign-in
  • Encrypted connection secrets with key rotation

Stop moving data by hand.

Start free with three projects, or talk to us about running it across your organisation.