llmist

Choose Your Path

Library Integrate llmist as a TypeScript library into your applications

CLI Run AI agents from the command line with powerful workflows

Testing Mock LLM responses for deterministic, fast tests

Core Capabilities

Home-Made Tool Calling

llmist uses its own block format for tools—no JSON mode or native function calling required. Works with any model.

Streaming Tool Execution

Gadgets execute the moment their block is parsed—not after the response completes. Real-time UX without buffering.

Multi-Provider Support

OpenAI, Anthropic, Gemini, and HuggingFace out of the box. Caching-aware, built-in cost calculation, multimodal support, and first-class reasoning model support.

Powerful Hook System

Three-layer architecture for observability, flow control, benchmarking, and deep app/UI integration.

Quick Start

npm install llmist

import { LLMist, Gadget, z } from 'llmist';

class FloppyDisk extends Gadget({
  description: 'Calculates how many 1.44MB floppy disks are needed to store a file',
  schema: z.object({
    filename: z.string(),
    megabytes: z.number().positive(),
  }),
}) {
  execute(params: this['params']): string {
    const { filename, megabytes } = params;
    const disks = Math.ceil(megabytes / 1.44);
    return `${filename} requires ${disks} floppy disk${disks > 1 ? 's' : ''}. Please label them 1 of ${disks}, 2 of ${disks}...`;
  }
}

const answer = await LLMist.createAgent()
  .withModel('sonnet')
  .withGadgets(FloppyDisk)
  .askAndCollect('How many floppies do I need for a 10MB file called DOOM.ZIP?');

console.log(answer); // "DOOM.ZIP requires 7 floppy disks..."

Packages

llmist

Core library with agents, gadgets, and providers.

npm install llmist

@llmist/cli

Command-line interface for quick completions and agent runs.

npx llmist agent "Your prompt"

@llmist/testing

Testing utilities and mocks for unit testing your agents.

npm install -D @llmist/testing