Hanzo Agent

Create, customize, and run autonomous agents

Build an agent right inside Hanzo — choose a model, write instructions, attach 260+ MCP tools, set a spending limit, or clone a persona. Or orchestrate multi-agent networks in code with hanzoai/agent across Rust, Python, Go, and TypeScript.

Create Agent Agent SDK

Create and manage agents in the web app at hanzo.ai/agent/create, or install the SDK (cargo add hanzo-agent · pip install hanzo-agent) from hanzoai/agent.

Available throughWeb AppRustPythonGoTypeScript260+ MCP toolsHanzo Cloud

Agents — Hanzo

api.hanzo.ai/v1/agentOpen in the console

Example project. Every name in it is one you would give your own.

Paste and ship

Paste this into any agent. It reads the skill manifest and calls Agent from there.

Agent prompt

Read https://hanzo.ai/skill.md and use Hanzo Agent in my project.

How Hanzo Agent works

From zero-code agent creator to multi-agent networks in Rust & Python

Every agent in your organization shares one identity, one tool catalog, and one budget ledger — whether you create it in the browser or deploy it from code.

Create in the Web App

Open hanzo.ai/agent/create to start from scratch, clone yourself from your writing, or install a preset persona. Pick any model on api.hanzo.ai, toggle MCP tools, and set a daily or monthly spending cap.

Put Agents in Any Room

Add your custom agents or the default Hanzo crew into any Chat, Dev run, or Team channel. Mention an agent by name or seat several in one conversation.

Pure-Rust Core & Multi-Language SDKs

Built on hanzo-agent (Rust) and the hanzoai/dev Rust harness for sandboxed execution, with first-class SDKs in Rust, Python, Go, and TypeScript (hanzoai/agent).

Capabilities

Everything an autonomous agent needs in production

Semantic & Rule Routing

SemanticRouter reads the request and picks the right specialist, or spreads turns across equals with load-balanced routing.

260+ MCP Tools & 125 Services

Every Hanzo MCP tool across files, AST, shell, browser, Kai, and Cloud services mounts over stdio or streamable HTTP.

Workflows & Shared State

Run steps in parallel, branch on conditions, and read or write one shared state object across every agent in a network.

Memory with Retrieval

A store, a retriever, and vector search over both, so an agent recalls the relevant part of a long history instead of replaying all of it.

Per-Agent Budgets & Tracing

Enforce daily, weekly, or monthly spending limits and per-task caps in micro-dollars, with full spans around every model and tool call.

Confidential TEE & Web3

hanzo-agent[tee] adds hardware attestation for confidential runs; [web3] adds wallets and on-chain identity so an agent can sign up and pay on its own.

Agent SDK · hanzoai/agent

Two agents and a semantic router

python

# Rust (cargo add hanzo-agent hanzo-agents) · Python (pip install hanzo-agent)

from agents import Agent
from agents.network import create_network, SemanticRouter

researcher = Agent(
    name="Researcher",
    instructions="You find and analyze information.",
    tools=[search_tool, analyze_tool],
)

writer = Agent(
    name="Writer",
    instructions="You create content based on research.",
    tools=[format_tool],
)

network = create_network(agents=[researcher, writer], router=SemanticRouter())

result = await network.run("Research and write about quantum computing")