hurozo Attend
First step

Set up your first agency

Create an agency, add workers, attach canon, and assign its first task.

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.

    Respond

    Answer from your own truth, not the open web.

    Agents guess and the wiki is stale. The Answer Desk pulls the answer from your canon, shows the exact line it used, and when your canon has no answer, it says so instead of inventing one.

    See it work → Book a demo
    The answer, on your data

    See the answer, and the line from your canon behind it.

    One question at a time, rotating. Each answered from your canon with the exact source shown, and the one it cannot answer yet, held.

    Attend · your support deskobserve mode · sample data
    QDo your jackets run true to size?
    Yes, our jackets are a regular fit, so take your usual size. If you are between sizes, we suggest sizing up for room to layer.
    from your canon
    Product guide · Outerwear fit"Regular fit. Between sizes, recommend sizing up."
    QHow long is your return window?
    You have 30 days from delivery to start a return, and refunds land within three to five working days of us receiving the item.
    from your canon
    Refund policy · §2"30-day return window. Refunds 3 to 5 working days after receipt."
    QDo you ship to Norway?
    Yes, we ship to Norway. Delivery is usually four to six working days, and any import fees are shown at checkout.
    from your canon
    Shipping · zones"Norway: 4 to 6 working days. Import fees shown at checkout."
    QIs the navy dye vegan?
    not in your canonIt does not guess. Held for a person. Your call to answer, or add it to canon.
    Answering from your canon · 1,240 answers, kept current
    pulled from your canoncited the linenothing from the open webheld when silentlogged

    A sample scan, in observe mode on sample data. The background behind the window is not real data. On your own tickets, every answer is pulled from your canon and cited to the source, and when your canon has no answer, the ticket is held for a person rather than guessed.

    How the Answer Desk works

    It answers from your truth, and shows its source.

    An Answerer, with the Logger writing down every step. It retrieves from your canon and your live systems, cites what it used, and holds when your canon has no answer.

    It answers from your canon, not the open web.
    Your policies, your product truth, your live systems. Never a confident guess pulled from somewhere on the internet.
    it read
    Product guide, outerwearyour canon
    Stock, liveyour systems
    It shows the line it used.
    Every answer cites its source, so your team can trust it and a customer can be told exactly where the answer came from.
    cited
    "Regular fit, size up"fit guide
    Traceable to the linegrounded
    When your canon is silent, it says so.
    It holds the ticket for a person instead of inventing an answer. The gap becomes a prompt to add to your canon, so the next customer is covered.
    your call
    Held, not guessed
    Flagged to add to canongap
    Why it matters

    One wrong, confident answer on the front page is the reputation you spend months rebuilding. Answers from your own truth, each with its source, are the ones you can stand behind.

    Questions

    What support leads ask about answering from canon.

    Where does the answer come from?
    Your canon: your policies, help center, product truth, and connected live systems. Not the open web, and not the model's memory.
    What if the answer is not in my canon?
    It says so and holds the ticket for a person. It does not guess. The gap is flagged so you can add the answer and cover the next customer.
    Can I see why it answered that way?
    Yes. Every answer cites the exact source line. You can trace any reply back to the canon it used, and so can your team.
    How is this different from a chatbot?
    A generic bot answers from the open web in a voice that is not yours. The Answer Desk answers only from your truth, shows the source, and holds when it is unsure.
    What keeps my canon current?
    The crew flags stale and missing answers from real tickets, and proposes updates for you to approve. Your canon sharpens as you run, it does not rot between edits.
    Is there a record of the sources?
    Yes. Every answer and the line it cited are logged, so any reply is auditable to its source. See Trust and security.
    Related jobs

    The answer is one station in the loop.

    Alongside · Respond
    First reply
    The whole reply drafted in your voice, held.
    →
    Before · Intake
    Morning triage
    The queue sorted before a word is answered.
    →
    After · Prove
    Prove every answer
    The logged answer and the source behind it.
    →
    Own your truth

    Own the truth your customers hear.

    Every answer from your canon, cited to the line, held when it is silent. The truth your customers hear is yours, and provable.

    See it work →Book a demo
    You own itBuilt for the EU AI ActYour data stays in the EUA person is always reachable