Cadenya
Paid
Developers building agentic workflows
Atizar
FreeOpen SourceAPISelf-HostedAgentic
Shipping an agent workflow to a client means the client will eventually click something they shouldn't — and without a review step baked into the execution path, the automation runs anyway.
Similar tools
JevProxyFreemiumThe proxy sits in front of your Cursor or Claude Code agent fleet and classifies incoming tool…
CadenyaPaidCadenya is an agent runtime built around a layered model: tools connect once through MCP…
Brain MemoryFreeBrain Memory stores agent decisions as Markdown files with YAML frontmatter, organized in a…
EmemFreemiumemem stores facts as short, signed tokens — each one a content-addressed handle that any agent…
Get This Tool
License: License: unverified
Local-run terms: Clone repo or install @atizar/* packages via npm; run with Docker for Postgres; full source available for modification and self-hosting.
Pricing
- Model
- Free
Summary
Atizar is an open-source, TypeScript-native framework for building agent workflows where humans stay in the loop before consequential actions execute. The core pattern: agents plan and gather, then pause for a sign-off before anything ships — emails send, records update, data moves. That approval gate is architectural, not bolted on after the fact. The self-hosted option means client deliveries stay off third-party infrastructure. Where it gets tight is documentation depth — the README carries most of the guidance, which means teams building complex branching logic are reading source code before long.
Bottom line: Pick Atizar when you're delivering an agent workflow to a client who needs to see every action before it fires — but expect to write your own abstractions the moment your branching logic outgrows what the existing examples demonstrate.
Find your fit
Before you choose Atizar
More options in Agent Frameworks. Compare pricing, platforms, and features side by side.
Brain Memory
Free · Open source
Long-term continuity in multi-session coding projects
Pros
- Human approval gates built into the execution model, so consequential actions — sending emails, updating records — cannot fire without a sign-off, which means you can hand this to a client without writing a separate audit wrapper.
- TypeScript-native agent code, so the workflow logic lives in the same codebase as the rest of your application — no context-switching to a separate DSL or canvas that generates code you didn't write.
- Self-hosted deployment option, so client data stays on infrastructure you control and you are not dependent on a third-party cloud runtime going down or changing its pricing.
- Open-source codebase, so when the docs run out — and they do run out — you can read what the framework actually does rather than waiting on a support ticket.
- API available, so the agent workflow is addressable from external systems, which means you can trigger automations from existing client tooling without rebuilding their stack around this framework.
Cons
- Documentation is thin beyond the README: teams building anything past the described use cases are reading source code to understand behavior, which adds days to scoping and slows onboarding for developers new to the project.
- No pre-built connectors or integration library is described in the repo or site — every SaaS connection your agent needs is a custom implementation, which means a five-integration workflow is five separate integration builds before you write a line of agent logic.
- The framework has no visual builder or canvas, so non-technical stakeholders cannot inspect or modify workflows without developer involvement; teams that need clients to configure their own automations will hit this wall immediately and typically move to a no-code-adjacent platform like n8n or Dify instead.
- Community size appears small based on available repo signals, which means when you encounter an edge case — and agent workflows generate edge cases reliably — there is precious little prior art to search before it becomes a support or debugging task you own entirely.
About
- Platforms
- Node.js, TypeScript, React (UI)
- API Available
- Yes
- Self-Hosted
- Yes
- Last Updated
- June 22, 2026
Best For
Who it's for
- Developers building trustworthy agent workflows
- Teams needing human oversight on actions
- Projects requiring TypeScript-native agent code
- Client deliveries with clean UIs
What it does well
- Human-approved email reply automation
- Client-facing workflow deployment
- Auditable agent-driven data processing
- Safe multi-step business automations
Integrations
Frequently Asked Questions
- Is Atizar free?
- Atizar offers a free option. Hosting, model usage, or optional services may have separate costs; check the pricing details and official terms.
- Is Atizar open source?
- Yes. Atizar is open source.
- Does Atizar have an API?
- Yes. Atizar exposes a developer API. See the official documentation at https://atizar.io for details.
- Can I self-host Atizar?
- Yes. Atizar supports self-hosting on your own infrastructure.
- What platforms does Atizar support?
- Atizar is available on: Node.js, TypeScript, React (UI).
Curated lists that include this category
Most agent frameworks hand you the execution loop and leave oversight as your problem. Atizar inverts that: human approval gates are part of the core workflow design, so agents that run tasks on their own — drafting emails, processing records, triggering downstream steps — pause and surface what they’re about to do before anything irreversible happens. The framework is TypeScript-native, built for developers who want agent logic that reads like production code rather than a visual canvas exported to YAML.
The differentiating feature is that human-gated execution is not a plugin or an optional middleware — the docs describe it as a first-class primitive in the execution model. For teams building client-facing automations where auditability matters (think finance, compliance, or any workflow where ‘the bot did it’ is not an acceptable answer), this changes the risk profile of shipping agents at all.
Atizar fits well when you need to hand off a working agent workflow to a non-technical client and guarantee they stay in control of consequential steps. It fits less well when you need rich pre-built connectors, a visual builder, or a managed cloud runtime — the vendor states none of those are in scope. Teams that need a drag-and-drop interface or out-of-the-box integrations with SaaS tools will be writing glue code from day one.
Installation is via npm, and the self-hosted model means there is no data leaving your infrastructure unless your agent integrations send it there. The open-source license means you can audit, fork, and extend — which community reports suggest is how most teams handle the gaps in built-in tooling.
Contribute
- Work on Atizar? Claim this listing.
- Embed a badge for your README or site.
Sign in to rate this tool or suggest an edit.