Activepieces
Build AI automations, flows and agents with governance built in
Linked from activepieces.com · checked 29 Oct
- Tool type
- Workflow automation and AI agent platform
- Made for
- Teams and developers automating recurring work
- Runs on
- Activepieces Cloud (Frankfurt) or your own Docker/Kubernetes servers
- Browser-based builder; self-hosted option; MCP server endpoint for external AI clients.
What it does
Activepieces is a workspace for building AI automations. Users describe a job in plain words and chat builds the flow or agent, or they assemble steps in a visual builder and add TypeScript where a step needs it. Flows, agents and tables share one workspace and can read and write the same data.
Teams use it for recurring work such as invoice intake, lead routing and support triage. Agents pick from connected apps and any MCP server, and sensitive steps can be gated for human approval. Projects, SSO, roles, audit logs and a bring-your-own secret manager provide governance, and the core is MIT-licensed open source.
It suits teams that want AI automation with access controls and the option to self-host with Docker or Kubernetes. Free includes 1,000 credits a month and Plus starts at $20 per month. AI steps use extra credits, and apps without a published piece connect through the HTTP piece or a webhook.
What it can do
- Build flows visually, with branching, retries, waitpoints, subflows, and human approvals
- Describe an outcome in chat and have flows or agents generated
- Run AI agents that pick their own connected tools and can gate sensitive steps for approval
- Store and query data in tables that flows and agents can read and write
- Expose workflows to Claude, Cursor and other MCP clients through an MCP server
- Self-host on Docker or Kubernetes, or use the managed cloud
Works with
- HubSpot
- Slack
- Stripe
- OpenAI
- Anthropic
- Google Gemini
- AWS Bedrock
- HashiCorp Vault
- Okta
- Microsoft Entra
Before you choose
- Each AI step uses 2 credits (fast model), 10 (smart model) or 20 (frontier model) on top of run credits; bringing your own AI keys on Plus and up lowers each AI step to 1 credit
- On Free, runs past the quota wait until credits refresh; on Plus and Team, overage is billed at $0.007 per credit
- The pricing page says its SOC 2 program is in progress
Listed under
Get started
Free plan has no card requirement; self-hosting is also available.
- Create an account on the cloud or install Activepieces with Docker or Kubernetes using the install guide
- Connect the apps your team uses once, so every flow and agent can reach them
- Describe a first job in chat or build a flow in the visual builder, then turn it on
Plans and prices
Prices in USDFree is $0 with no card required. Credits reset daily on Free and monthly on Plus and Team. Credit overage on Plus and Team is $0.007 per credit. Embed is listed from $36k a year via sales. Ultimate pricing is custom via sales.
- 1,000 credits a month
- Unlimited flows
- Up to 5 users
- 10,000 credits a month
- Unlimited flows
- Up to 5 users
- Bring your own AI keys
- 50,000 credits a month
- Unlimited projects
- 25 users included
- SSO and standard roles
Source: activepieces.com/pricing, checked 29 Oct 2026. Prices can change; check before you buy.
Sources for this overview
Product pages, documentation, and pricing checked 29 Oct 2026. Each source supports part of the overview; availability and prices can change.
- Activepieces: unify your team's AI ↗www.activepieces.com
- Welcome to Activepieces - Activepieces ↗www.activepieces.com
- Pricing | Activepieces ↗www.activepieces.com
Ratings and reviews
No ratings for Activepieces yet.
What’s changed
Free accountSignals
Free accountData and security
Free accountCompare with similar tools
Free accountWhere this comes from
How we research- Company website
- activepieces.com: product, pricing and policy pages
- Hacker News
- Launch posts and discussion
- YC directory mirror
- Batch and company name
- Social accounts
- Accounts the company links, and its posts on X
- Human review
- Checked by a person
- knowledge_base
- manual_review
Last checked 29 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.