Available now: version 0.2.0, with the local dashboard

Build agents you trust

Connect your tools. Set the rules. Build and run AI agent teams with built-in guardrails, human approvals and complete visibility. Free, open source and self-hosted.

npx agentbuildkit

Apache 2.0. macOS and Linux, Node.js 18 or newer. The demo and dashboard need no API key.

$ agentkit demo --approveReal output
[4] TOOL BOUNDARY
    tool: cancel_ticket
    result: rejected
[5] SIDE EFFECT
    tool: schedule_visit
    approval: required
✓ Approved
[7] EVALUATION result: PASS

Local dashboard with demo data. Account A-100 is a fixture.

Works with OpenAI built inAnthropic built inMicrosoft Foundry optional GitHub built inCursor built inMCP servers built inHTTP APIs genericGoogle Sheets via Composio, read-only, optional

Connected apps through Composio. Google Sheets today, read-only. Each connected tool has to be explicitly approved, is pinned to a version and only reads, so more apps can be added one at a time. More are on the roadmap.

Available now

One command. Running in minutes.

It checks your system, installs AgentBuildKit and both templates in their own environment, checks they work, then offers the dashboard and the offline demo.

You need

macOS or Linux and Node.js 18 or newer. Windows hasn’t been verified for the dashboard.

It installs

The agentbuildkit 0.2.0 harness, the agentbuildkit-templates 0.2.0 package and the agentkit command, using uv. Python 3.12 is set up for you.

Where

An isolated uv tool environment. It never changes your system Python. If uv is missing, it asks first.

API key

Not needed for the demo or the dashboard. For live runs, set OPENAI_API_KEY or ANTHROPIC_API_KEY.

Dashboard

agentkit dashboard runs on 127.0.0.1 for one local operator, with no login. Opening it never calls a model.

Remove it

uv tool uninstall agentbuildkit

$ npx agentbuildkit --no-demo --no-dashboardInstaller output, npm 0.0.4
AgentBuildKit
Build Agents You Trust
 
Checking your system...
✓ Linux x64 detected
✓ uv available
✓ AgentBuildKit installed
 
AgentBuildKit Doctor
Python             3.12.13          ✓
AgentBuildKit      0.2.0            ✓
SQLite             3.53.1           ✓
Templates          installed       ✓
Startup lock       POSIX flock     ✓
 
OpenAI             not configured
Anthropic          not configured
Status: ready for offline demo
 
Dashboard skipped. Run `agentkit dashboard --state-directory DIRECTORY` when you want it.
Demo skipped. Run `agentkit demo` when you want it.
New in 0.2.0

Every run, every approval, in view

The local dashboard shows what your team did, what it’s waiting for, and what needs a check. Approve or reject changes from the browser. It reads what was recorded and never invents a step.

Waiting for approval
schedule_visit for A-100
Proposed byTaylorWhat runsschedule_servicePaused22 Nov 2026, 20:38 UTC
If you approve
Sent at most once. Never replayed.
If you reject
Nothing is sent.
Approve and sendReject

Approve, with the facts in front of you

Demo data

The action, the customer and what happens either way. Nothing is sent until you decide.

Completed
Alex routed to Taylor. Taylor read the service record. Jordan approved the draft. You approved. Completion recorded.

See the route each request took

Demo data

Steps the run recorded get a blue ring. Agents it didn’t use stay faint.

20:38:03Agent run Alex, gpt-4.1-miniCompleted
20:38:03Tool call Taylor, read_serviceCompleted
20:38:03Paused for approvalPending
20:40:12Approval acceptedCompleted
20:40:12Side effect completedCompleted

An audit trail you can read

Demo data

Every agent run, tool call, pause and approval, with the provider and model. Hashes, never argument values.

Screens use the shipped demo data. One local operator, bound to 127.0.0.1. Opening the dashboard doesn’t call a model.

Available now

Start from a working team

Two templates ship with agentbuildkit-templates. Create your own instance, edit its draft, evaluate it offline, then activate it.

Customer Service

Approvals on

Alex classifies a request. Python routes it to Sam or Taylor. Jordan reviews the reply. A service change pauses for your approval.

AAlex, managerSSam, accountsTTaylor, serviceJJordan, reviewer

Operations Reporting

Read-only

An analyst reads the jobs and writes a summary. A reviewer checks it names only those jobs. No approval actions, nothing written back.

AAnalystRReviewer
  1. 1Create teamAn inactive instance of a template.
  2. 2Edit the draftVersioned. Nothing goes live.
  3. 3ValidateChecked against the template.
  4. 4Evaluate offlineScripted suite, no API key.
  5. 5ActivateExplicit. Rollback is too.

Draft teams from Claude or Cursor

A local MCP server can create an inactive team, edit its draft, validate it and run the scripted evaluation. It can’t activate, run, approve or connect anything: that stays with you in the dashboard.

Available now
Available now

Guardrails are part of the architecture

The production layer around your agents, built on Microsoft Agent Framework. Plain Python, not the model, decides which tools an agent may use and whether a side effect may run.

Agent teams

Typed agents with clear roles. An orchestrator routes work to specialists and a reviewer checks the result.

Your model, per agent

OpenAI and Anthropic built in. One workflow can mix them without branching on the provider.

Deterministic workflows

Plain Python decides the route, not the model. Unknown routes fail closed.

Tool allowlists

Each agent gets only the tools you grant. Anything else is rejected and logged.

Human approval

Reads run. Side effects wait for a person, and an approved action runs at most once in the local store.

Durable state and audit

Runs pause and resume from SQLite. Every decision is recorded with its provider and model. OpenTelemetry export included.

Offline evaluation

Scripted suites check the route, tools and approvals against expectations, with no API key.

Tools and connectors

GitHub, Cursor, MCP servers and your HTTP APIs, each behind a permission. Optional read-only Google Sheets via Composio.

A CLI that tells you the truth

agentkit doctor checks Python, SQLite, templates and which providers are configured.

One real run of agentkit demo --approve
9 checks
Route chosen by Python1
Allowed tool run1
Unlisted tool rejected1
Side effect paused1
Approved, ran once1
Schema validations3
Evaluation passed1

Your keys, state and audit records stay in your environment. Only the prompts you send to your chosen model leave it.

Coming next: not in 0.2.0

Where AgentBuildKit is going

The framework stays free. These are planned, not shipped, and none has a date or a price.

Watch releases on GitHub

Questions

Is it free?

Yes. AgentBuildKit is open source under the Apache 2.0 licence. You pay only your model provider, and only for live runs.

Do I need an API key?

Not for the offline demo, the dashboard or the evaluation suites. Live runs need an OpenAI or Anthropic key for the agents that use them.

Which models does it support?

OpenAI and Anthropic are built in, and one team can mix them. Microsoft Foundry is an optional extra.

Where does it run?

On your machine. The dashboard binds to 127.0.0.1 for one local operator. It isn’t a hosted or multi-user service.

Does it work on Windows?

The bootstrapper supports macOS and Linux. Windows hasn’t been verified for the dashboard or the MCP server.

Can I build my own team?

Today you create instances of the two shipped templates, edit their drafts, evaluate them offline and activate them. You can also build agents directly in Python with the harness.

Does approval guarantee the action happens once?

Approval runs an action at most once in the local store. An interrupted action stays unresolved and is never replayed. It isn’t exactly-once delivery to an outside system.

What is it built on?

Microsoft Agent Framework. AgentBuildKit adds the guardrails, approvals, state, audit and evaluation around it.

Install it now

Run the demo and open the dashboard in a few minutes. No account, no API key, nothing to pay.

npx agentbuildkit
View the source on GitHub