mirror of https://github.com/hanzoai/python-sdk.git synced 2026-10-23 20:23:00 +00:00

Files

T

zeekay 241ca6dc59 the front door tells the truth about the client behind it

README: PyPI serves 3.2.1 and this tree is 3.2.11, so the install line is the
git one. Between the two the default version left the operation ids — 3.2.1
carries get_v1_keys/get_v1_tools (182 api modules, 2172 models) and this carries
get_keys/get_tools — so `pip install hanzoai` above a `get_keys()` quickstart
hands the reader a client that cannot run it. Counts are measured off the pinned
document: 1814 paths, 2479 operations, 191 tags + default_api = 192 API classes
over 2460 models, 2659 importable modules.

Auth was wrong everywhere and it was the same root cause: the document declares
no securityScheme, so every operation got an empty auth_settings and
Configuration(access_token=…) is read by nothing. Serializing get_keys from such
a Configuration produces a request with no Authorization header at all — the
five flows have been calling anonymously and reading the refusal as a bad key.
examples/client.py now passes the bearer through ApiClient's header_name/value
pair, verified against a local server that echoes what it received, and a smoke
test pins both halves so the day cloud emits the scheme, the test fails and the
header moves back into Configuration.

The examples gate resolved class names and stopped there, which is how `money`
and `tools` once passed it while every call in them named a method that no
longer existed. It now walks the syntax tree for attribute accesses on anything
bound to a *Api and asks the class: ten names resolve, and a stale get_v1_tools
fails in the same second the import passes.

Deleted five example scripts that imported hanzoai.Hanzo, hanzoai.cluster,
hanzoai.agents and hanzoai.completion — none of which exist — or were dict
literals captioned "in practice this would be called through the MCP interface";
compileall passed all five, because syntax is not a claim. Deleted three docs
that were fabrication rather than drift and that nothing rendered: FEATURES.md,
QUICKSTART.md, GPT5_ORCHESTRATION.md.

pkg/hanzo-agent said `pip install hanzoai` and `from hanzoai import Agent,
Runner`; the client has no Agent. Its own `full` extra resolved to
hanzoai[web3,tee,marketplace,cli], extras the client does not declare, so it
installed the client and none of the extensions. pyproject said BSD-3-Clause
while LICENSE is the Apache 2.0 text. test_smoke imported tracker_api and no
/v1/tracker path is emitted any more, so the suite carried a failing test naming
a module the client does not have; it reads analytics_api now, off the client.

2026-08-14 14:31:08 -07:00

4.7 KiB

Raw Blame History

Hanzo Python SDK

A workspace of 65 packages. Two things live here that people usually arrive for: the API client, and the MCP server.

The API client (`hanzoai`)

Generated from the Hanzo API's own OpenAPI document — 1814 paths, 2479 operations, 192 API classes over 2460 models. It is documented in the repository README, which carries the install line, the quickstart and the five runnable flows in examples/.

import os
from hanzoai.cloud import ApiClient, Configuration, KeysApi

client = ApiClient(
    Configuration(host="https://api.hanzo.ai"),
    "Authorization", f"Bearer {os.environ['HANZO_API_KEY']}",
)

with client as api:
    for key in KeysApi(api).get_keys().keys or []:
        print(key.prefix, key.type, key.created_at)

The rest of this site is the MCP server, the agent framework and the tool packages.

Installation

=== "MCP server" bash pip install hanzo-mcp

=== "Agent SDK" bash pip install hanzo-agent

=== "Using uv" bash uv pip install hanzo-mcp

Quick Start

Using with Claude Code

The easiest way to use Hanzo MCP is with Claude Code:

# Install globally
uvx hanzo-mcp

# Or run directly
uvx hanzo-mcp --transport stdio

Agent SDK Example

from agents import Agent, Runner

agent = Agent(
    name="assistant",
    instructions="You are a helpful assistant."
)

result = Runner.run_sync(agent, "Hello!")
print(result.final_output)

MCP Tools Example

from hanzo_mcp import create_mcp_server

# Create server with all tools
server = create_mcp_server()
server.run()

Packages

Package	Description	PyPI
`hanzo-mcp`	MCP server with all tools
`hanzo-agent`	Agent SDK
`hanzo-tools-shell`	Shell/command tools
`hanzo-tools-fs`	File system tools
`hanzo-tools-browser`	Browser automation
`hanzo-tools-memory`	Memory/knowledge tools

Features

30+ MCP Tools

File System: read, write, edit, tree, find, search, ast
Shell: cmd, zsh, bash, ps, npx, uvx, open
Browser: Full Playwright automation with 70+ actions
Memory: Unified memory tool with recall, create, update, delete
Reasoning: think, critic for structured AI reasoning
LSP: Go-to-definition, find references, rename, hover
Refactor: Advanced code refactoring with AST support
Agent: Run external AI agents (Claude, Gemini, Codex, etc.)
LLM: Direct LLM access and multi-model consensus

Auto-Backgrounding

Long-running commands automatically background after 30 seconds:

# This will auto-background if it takes too long
cmd("npm install")

# Check status
ps()  # List all background processes
ps(logs="cmd_xxx")  # View output
ps(kill="cmd_xxx")  # Stop process

Multi-Backend Support

Use the MCP with any AI assistant:

Claude Code - uvx hanzo-mcp
VS Code/Cursor - Hanzo extension with auto-detection
Antigravity - Full integration via Open VSX
Any MCP Client - Standard MCP protocol

Documentation

Getting Started

Reference

Tools Reference - All 30+ tool packages
Agent Reference - Agent SDK documentation
MCP Reference - MCP server documentation
Core Libraries - Supporting packages (async, consensus, network)

Tool Categories

Filesystem Tools - File operations and AST analysis
Shell Tools - Command execution and process management
Browser Tool - Playwright automation
Memory Tools - Persistent memory and knowledge bases
Reasoning Tools - Structured thinking and analysis

4.7 KiB Raw Blame History