OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Frameworks

Axar

Axar is a lightweight, Apache-2.0-licensed TypeScript framework for developing LLM-powered agentic applications on Node.js and server environments. It provides class-based agents, model and system-prompt decorators, typed inputs and outputs, schema validation, tool integration, dynamic prompts, dependency injection, streamed responses, logging and monitoring. The repository documents integrations with OpenAI, Anthrop

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Axar — OpenViST project profile in Agent Frameworks.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Axar when a Node.js or TypeScript project needs code-defined LLM agents and workflows with explicit behavior, typed structures and model-provider flexibility.

Who it is for

  • TypeScript and JavaScript developers building LLM-powered applications
  • Teams developing production-oriented agentic workflows
  • Developers who want explicit code-level control over prompts, tools, schemas and validation

Documented uses

  • Define an LLM agent with a selected model and system prompt
  • Build workflow-oriented agent applications with typed inputs and outputs
  • Add tools and dynamic prompts to an agent
  • Return structured, validated responses
  • Stream model output and monitor agent operations
Before you choose

Requirements & limits

The repository is public and licensed under Apache-2.0. The documented setup uses npm packages and ts-node; the OpenAI example requires an OpenAI API key.

  • TypeScript is the required implementation language in the documented examples.
  • The framework uses experimental decorators, and the repository states that ts-node is currently required for the shown workflow because tsx does not fully support them.
  • Typed schemas and validation are available for agent outputs and tool parameters.
  • The documented model integrations include OpenAI, Anthropic and Gemini.
  • The framework is designed around explicit code and minimal abstraction rather than a visual builder.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up a first Axar TypeScript agent

Prepare a TypeScript project and create the documented OpenAI-backed example.

Before you start

  • Node.js and npm tooling
  • TypeScript
  • The @axarai/axar, ts-node and typescript packages
  • An OpenAI API key for the documented model example

Steps

  1. Create a project directory, initialize it with npm and install @axarai/axar, ts-node and typescript.

  2. Generate a TypeScript configuration and set strict mode, CommonJS modules, an ES2020 target, esModuleInterop, experimental decorators and emitDecoratorMetadata.

  3. Define a TypeScript Agent class with the documented model and system-prompt decorators.

  4. Call the agent’s run method with a text prompt, set the OPENAI_API_KEY environment variable and invoke the file through ts-node.

Expected outcome

A configured TypeScript project containing the documented minimal agent example and its execution command.

Guide limitations

  • The repository notes that ts-node is currently required because tsx does not fully support the experimental decorators used by Axar.
  • The example uses OpenAI and therefore requires the corresponding API key.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
3 October 2026
Original website title
axar-ai/axar · Official homepage title · retained in its source language
How we review sourcesSuggest a correction