Search

More Filters

Show More
Add Listing
Go Back

experts

0
New
Multi-Agent Platforms & Frameworks
Report

Report Abuse

Basic Information

Experts.js is a Node library and framework for building Multi AI Agent Systems using OpenAI's Assistants API. It provides high-level abstractions for creating Assistants, Tools, and Threads so developers can compose multiple specialized assistants into a Panel of Experts architecture. The library hides low-level Run and Run Step management and exposes a simple ask() interface to send messages to assistants within managed threads. It supports ES6 imports and CommonJS require, installs via npm, defaults to the gpt-4o-mini model unless overridden, and integrates OpenAI streaming events, vector store file search, and tool/function calling. The README includes examples such as a product catalog, streaming responses from Express, and token usage metrics. The project also documents deployment patterns, environment variables for model configuration and debugging, and development setup using dev containers, making it a practical toolkit for developers who want to orchestrate multiple LLM-backed agents and related resources.

Links

Stars
1058
App URL
https://github.com/metaskills/experts
Github Repository
https://github.com/metaskills/experts/blob/main/README.md

Categorization

App Details

Features
High-level Assistant abstraction that maps to OpenAI Assistants API and accepts name, instructions, model, tools, and run options. Tools are implemented as Assistant subclasses and can be attached via addAssistantTool to create nested agent capabilities. Thread management creates and stores per-agent threads to serve as managed context windows and avoid thread locking issues. Simple ask() API accepts strings or native message objects and hides Run management. Streaming and event support surface server-sent events like textDelta, imageFileDone, toolCallDone, runStepDone, with async variants for post-run work. Vector store integration and file_search tool_resources are supported for large-file search. Non-LLM tools are supported by setting llm:false and implementing ask(). Advanced hooks beforeInit and afterInit allow lazy resource creation. Deployment options include using assistant ids and skipUpdate, and debug toggles are provided via environment variables.
Use Cases
Experts.js reduces the engineering effort required to build multi-agent applications by encapsulating common Assistant and tool orchestration patterns. Developers do not need to manually manage Runs or Run Steps, threads, or event streaming, which simplifies building chat apps, search-backed assistants, and multi-tool workflows. The framework enforces a reusable tool interface so specialized assistants can be composed, redirected, or returned as parent outputs, enabling complex pipelines like product catalogs and OpenSearch-backed tools. Streaming support enables real-time client responses such as Express streaming examples. Built-in thread management prevents concurrency and locking problems. Vector store and file search support make retrieval-augmented interactions practical. Hooks and configuration options support lifecycle management and production deployments, while debug and token-usage metrics help monitor and optimize cost and performance.

Categories

Content Creation (5042)
Monetization & Business (2246)
Development Tools (1407)
Marketing & Social Media (1403)
Education & Career (1380)
Multi-Agent Platforms & Frameworks (1009)
Health & Life (957)
AI Resources (920)
Agents & Automation (778)
Data & Analytics (764)
Productivity Tools (679)
Writing & Editing (666)

Similar Listings

listing image

virtual lab

0 (0 review)
listing image

mnemo

0 (0 review)
listing image

EdgeChains

0 (0 review)
listing image

RAG Agents Accelerator

0 (0 review)
listing image

LLM Zero to Hundred

0 (0 review)

Featured Listings

Tags

Please fill the required fields*