namzu.ai
Integration · @namzu/anthropic

Namzu × Anthropic. Frontier model API behind Claude.

The @namzu/anthropic package wires Anthropic's Messages API into Namzu's LLMProvider interface. Tool calling, streaming, structured outputs, and stop sequences are mapped one-to-one with no abstraction tax.

01 · Install
$ pnpm add @namzu/sdk @namzu/anthropic
02 · Why pair the kernel with Anthropic

Claude's tool-use API is the closest first-party fit to a kernel-style runtime — explicit tool blocks, deterministic stop reasons, and a stable schema across model generations. Namzu surfaces those primitives as kernel signals rather than burying them in a framework abstraction.

  • Tool calling and tool-result handling without DSL translation
  • Streaming responses surfaced as kernel events for IPC and persistence
  • Per-agent model + temperature config via the kernel's provider scope
  • Vision and document inputs passed through unchanged
03 · Example
import { createKernel } from '@namzu/sdk'
import { anthropic } from '@namzu/anthropic'

const kernel = createKernel({
  provider: anthropic({
    apiKey: process.env.ANTHROPIC_API_KEY!,
    model: 'claude-sonnet-4-6',
  }),
})

const agent = kernel.spawn({
  systemPrompt: 'You are a careful research assistant.',
  tools: [searchWeb, summariseDocument],
})

await agent.run('Find the latest paper on agent kernels and summarise it.')

Ship Anthropic agents on a runtime you can own.

Anthropic owns the model. Namzu owns the runtime. The provider package is a thin adapter, not a wrapper, so you keep everything Anthropic ships and add the kernel concerns you do not get from the model API alone.