Attend Attend

Worker Tools

Manage worker

Tune the role and tools for this agency teammate. Graph wiring stays internal.

Tools

Loading...

Worker View

No tools selected
Team worker Untitled Agent Describe what this agent should do. HurozoAgent

Worker Chat

Describe the worker you want to add to this team, or what you want to change.
Draft JSON 
{}

Teams

Teams are managed from Agency.

Connectors

Connectors

Connect services for the current agency

Loading connectors…

Agency

Team workspace

Planning

Filters

Build

Runs

Execution activity

Runs

Inspect Worker work, tool calls, reasoning summaries, and failures.

Select a run to inspect its activity.

Files

Agency files

    Documentation

    Contact Us

    New message

    To: hi@hurozo.com

    Hurozo CLI

    Terminal-first agent work

    Hurozo CLI

    The same Hurozo agent runtime, model routes, approvals, and connected tools from your shell.

    Install with Homebrew 
    brew tap hurozo/hurozo-cli
brew install hurozo-cli

    View latest GitHub release

    Hurozo CLI running in a terminal

    Model routing

    Switch between Hurozo, Gemini, Opus, and ChatGPT-backed routes when your account is configured for them.

    Approvals

    Review tool permissions from the terminal and keep remembered decisions scoped to the exact operation.

    Second opinions

    Use model arbitration to ask another available route for critique, checks, or a different perspective.

    Connected tools

    Use the same account-backed connectors and skills available in Hurozo's desktop and WebCLI experiences.

    First run

    After installation, run hurozo. The CLI walks you through account authentication and any model-specific sign-in or secret setup that your selected route requires.

    Hurozo SDK

    Python agent invocation

    Hurozo SDK

    Invoke saved Hurozo agents from Python using the same API tokens, inputs, outputs, and runtime routing used by Agent Builder.

    Install 
    pip install hurozo
    getting_started.pyPython
    import json
import os

from hurozo import Agent

os.environ["HUROZO_API_TOKEN"] = "YOUR_API_TOKEN"
agent = Agent("sentiment-analyser")
inputs = {
    "prompt": "Analyze sentiment and return JSON. Text:",
    "input": "I am very happy!",
}

result = agent.run(inputs)
print(json.dumps(result, indent=2))

    Agent lookup

    Use an agent name, saved agent id, or agent UUID. The SDK resolves the correct execution target.

    Structured inputs

    Pass a Python dictionary matching the API input keys exposed by the saved workflow.

    Hosted runtime

    Runs against Hurozo's execution endpoint, including model routing, tools, outputs, and usage accounting.

    Token setup

    Create an API token from Account settings -> Manage API Tokens, then set it as HUROZO_API_TOKEN before running scripts.

    Billing and charges · a live case

    “What’s this charge?” Explained from your billing, before it becomes a dispute.

    The question that turns into a chargeback when no one answers it. Attend reads the line on your statement, explains it in plain words from your billing system, and holds any correction for a person to approve.

    See it work → Book a demo
    New message · email 09:14
    There are extras on our March statement we don’t recognise, about €245. What are they?
    Attend is reading it
    What a confused customer costs you

    Right now, an unanswered charge becomes a chargeback.

    The customer cannot tell what a line means, cannot get a fast answer, and goes to their bank instead. By then it is a dispute, with fees, and your team is arguing a charge that was probably correct.

    48%
    of shoppers have disputed a charge they later found was legitimate. Confusion, not fraud.
    Chargeback research · 2026
    €128
    the all-in cost to you of a single chargeback, more than many disputed items are worth.
    Mastercard / Javelin · 2025
    50%
    of cardholders now go straight to their bank instead of asking you, believing it is faster.
    Chargeback Field Report · 2024

    External research, cited above. Figures are card-industry data, in euros for illustration.

    How the crew handles this one

    It explains the line first, and holds any correction for a person.

    Attend · billingobserve mode · sample data
    QueryThere are extras on our March statement we don’t recognise, about €245. What are they?
    Statement MAR-2026 · Northwind Fleet · 14 vehicles
    Monthly plan14 vehicles
    €1,120
    Winter service packMarch
    €155
    Per-vehicle add2 vehicles
    €90
    The €155 is the winter service pack, added when you renewed in January. That one is correct and in your plan. The €90 is a per-vehicle add for 2 vehicles. from your billing + canonStatement line · your plan · billing policy
    correction heldThose 2 vehicles were returned on Feb 27, so the €90 is charged in error. A €90 correction is held for a person to approve.
    billing + plan, liveStatement · billing system · your plan

    Sample data, in observe mode. On your own statements, every charge is explained from the billing line and cited, and any correction is held for a person to approve.

    You already run a billing platform

    Your billing platform shows the charge. It doesn’t explain it.

    Your billing platform
    It presents the statement and takes the payment. But it shows a line code, not an answer. It cannot tell your customer what a charge means, or catch the one that is wrong.
    A general chatbot
    off-the-shelf AI
    Fast to set up. It guesses at a charge it cannot see, or waves through a correction it should never make on its own.
    By hand
    your team + the billing system
    Real judgment. But it is slow, the customer has already called their bank, and a correct charge turns into a dispute.
    Attend, on this job
    Reads the line on your statement, explains it in your voice, and holds any correction for a person to approve. Keeps the billing platform you already run.
    What it grounds in for this job

    It reads your statement, your billing system, and your plan. No charge explained it cannot point to.

    Your canonBilling and plan policy Billing and statementsThe live statement, from your billing system Your planWhat each customer actually signed up for

    Attend connects to the systems you already run. See the integrations →

    Proof

    Every charge explained, and the line behind it, is yours to inspect.

    traced to your billingsample log · STM-MAR-2026 · 09:03 CET
    Traced from: the statement + your billing system + your plan. Held corrections are logged too.
    read the statementtraced to the billing linenothing from the open weblogged and replayable

    Every explanation, and every held correction, is one record you can open, replay, and export. Designed around the EU AI Act, on a source you own.

    You own it EU AI Act, by design EU residency on our roadmap A person approves every correction
    Own the answer before it becomes a dispute

    Answer “what’s this charge?” from your billing, every time.

    Explained from your statement, traced to the line, and a person on every correction.

    Book a demoSee it work →

    The pilot is risk-free on the investment. If it does not hit the KPIs you agree, you do not pay for the sprint.

    See how the first weeks work: First steps →