Openai agents

Braid is a TypeScript DAG runtime for LLM agent orchestration — parallel, conditionally-routed graphs of isolated model invocations with per-node model control. OpenAI-compatible, framework-agnostic, zero dependencies.

Run local LLMs on limited VRAM safely. Dynamic model sharing and OpenAI API drop-in replacement.

Universal multi-IDE agent teamwork skills: one team-orchestrator skill + per-tool role configs for 12 coding tools (Claude, Codex, Copilot, Cursor, Windsurf, Trae, Qoder, Antigravity…)

Agents you can replay, fork and trust. An agent framework for TypeScript and Python, built on an append-only event log.

AI-powered autonomous agent that controls computers through visual understanding and task planning

The coding agent built for local models. One native binary for your terminal that finds the model your local server has loaded, and tells you whether that model can actually drive tools.

Local-first AI Agent framework. DeepSeek-compatible (deepseek-v4-pro / deepseek-v4-flash), multi-model (OpenAI/Claude/GLM/Xfyun), Core Loop Kernel, sub-agent multimodal delegation, Windows desktop app.......本地优先的 AI Agent 框架，DeepSeek 兼容。

Nessy — an AI agent harness framework for Java

One desk where your AI agents live and work: your folders, your keys, your models. Schedules, memory, plugins, IDE-style UI. Fair-code.

The best free desktop app for autonomous AI agents & AI employees. Full computer use & browser use on macOS, Windows, & Linux. Operates logged-in web sessions, clicks, types, moves files, & runs CLI commands. Secure, production-grade local runtime for multi-agent LLM orchestration (Anthropic Computer Use, OpenAI Operator & Browser Use alternative).

Self-hosted starting point for agentic projects: agents built from skills and tools, a storage library over your own tables, a multi-agent workflow, and a client portal. Skills and tools are database rows, editable with no redeploy.

Embodied AI assistant for the Reachy Mini robot — real-time voice interaction, 30Hz presence loop, smart home control, and multi-user memory, optimized for ultra-low latency

MCP server exposing the full OpenAI API (Responses, Chat, Images, Audio, Embed, Files, Batch, Vector Stores, Fine-tune, Uploads) — 42 tools for Claude Code, Cursor, and any MCP client.

Real-time cost enforcement for AI agent sessions — works with OpenAI and Anthropic

Distributed multi-machine AI agent team platform

AI Agent Network CLI — Local-first multi-agent orchestration across 6 runtimes (Claude Code CLI / Claude Agent SDK / Codex SDK / Codex app-server / Grok Build ACP / OpenCode CLI) and 8+ LLM providers. Apache 2.0.

Browser-adapted MCP TypeScript SDK - Dynamic tool registration for W3C Web Model Context API and AI agent integration

Tiny, extensible AI agent.

Transparent AI agent tracing

AI agent runtime with WebSocket protocol, artifact handling, and secret management

AI agent SDK for building intelligent assistants - tree-shakable and comprehensive

The TypeScript-native AI agent framework with multi-provider LLM support

Multimodal preflight for any AI agent — AI SDK middleware for transparent image optimization

AI agent observability and control from your iPhone or Apple Watch

TITAN — the AI agent that provably teaches itself. TypeScript agent framework with replay-verified self-learned skills (Muscle Memory), one-minute Welcome Mode setup, a living desk with heartbeat/reminders/proactive memory, multi-agent orchestration, 36-p

Unified AI Agent Library - Supports Claude Agent SDK, Codex SDK, and proxy providers (OpenAI, Gemini, Z.AI)

Self-hosted AI agent platform — autonomous, multi-channel, multi-model.

Framework-agnostic AI Agent SDK with tool calling, conversation management, and automatic summarization

23 AI agent tools + context-compaction helper as standalone Vercel AI SDK modules

Complete AI agent implementation with unified core and tools functionality - conversation management, plugin system, and advanced agent features

AI agent platform — security-first orchestration for autonomous agents

Tool of Dev — AI agent in your terminal

AI agent CLI and library for Node.js — streaming, multi-provider, tool execution, autonomous tasks

Ethos — TypeScript AI agent framework where personality is architecture. Curated toolsets, first-person identities, scoped memory.

🐟 AZERCLAW — Open-source, BYOK AI agent for your terminal. Your AI, Your Keys, Your Way.

OpenAI LLM provider (ILlm) for @mcp-abap-adt/llm-agent.

DeepSeek LLM provider (ILlm, extends OpenAIProvider) for @mcp-abap-adt/llm-agent.

OpenAI embedding provider (IEmbedderBatch) for @mcp-abap-adt/llm-agent.

Lightweight multi-LLM agent library for building CLI AI assistants

Headless, multi-provider LLM agent engine for workflow automation. Pause/resume, MCP, skills, R2/Workers compatible.

OpenAI LLM Agent Interface

Multi-provider LLM agent runner with tool plugin execution (Anthropic, OpenAI, mock). Used by Studio.

The llm-agent is a TypeScript project designed to facilitate interactions with language models from multiple providers. It supports various LLM API providers that are compatible with the OpenAI API structure, allowing users to manage chat interactions and

Ollama LLM provider (ILlm, extends OpenAIProvider — Ollama's OpenAI-compatible /v1 API) for @mcp-abap-adt/llm-agent.

Improve, rewrite, debug, evaluate, and compress prompts for any LLM agent. Ships an installable Skill plus a deterministic local CLI.

Quickly scaffold a TypeScript-based LLM agent application with local or remote AI models

OpenAI integration for computer-agents SDK. LLM agent runtime and OpenAI API support.

Public MCP server for Ordering.Tools — discover venues, read menus, prepare orders. Used by Claude Desktop, Claude Code, OpenAI Apps in ChatGPT, and any MCP-compatible LLM agent.

TypeScript LLM agent framework: ReAct loop, tool calling, MCP, multi-agent workflows, memory, guardrails. Anthropic Claude, OpenAI, Ollama via fetch — vendor-neutral, zero SDK deps.

一个轻量、高效、模型无关的 LLM Agent 框架，采用 ReAct 单调用循环架构

Provider-agnostic LLM agent framework. One LLMProvider interface, two adapters (Anthropic + OpenAI-compatible), tool loop, router/specialist orchestration, streaming, confirmation gate, typed errors, retry, channels. Zero runtime dependencies.

开箱即用的 LLM Agent SDK — 配个 API Key 就能跑

Enforce real-time token budgets and spending limits for OpenAI, Anthropic Claude, and Google Gemini API calls in Node.js

Standalone prompt agent powered by OpenAI SDK.

Local AI assistant: 80 tools (Gmail, Calendar, Drive, GitHub, Slack, browser, code, files), 38 agents, visual workflows (Studio, AWF, WebCraft). Install with `npm i -g nothumanallowed`, run with `nha ui`. Free tier built-in (Liara), no API key required. Y

MCP server that lets any LLM agent (Claude Desktop / Cursor / OpenAI Agents SDK) buy products on JPYC EC Platform via x402 — gasless EIP-3009 payments over HTTP.

Lightweight, zero-dependency LLM API cost & token usage tracker for OpenAI, Anthropic, Gemini, Mistral, Groq, and DeepSeek

AI runtime for agent-driven products — agents, memory, LLM providers (Inference Snaps, Ollama, OpenAI-compatible), tools, and orchestration. Anthropic-SDK-free.

Smart LLM router — save 84% on inference costs. 78 models (6 free). Pay by credit card (API key) or USDC (x402 wallet).

TES SDK — LLM observability and lifecycle tracking via Pentatonic Thing Event System. Track token usage, tool calls, and conversations. Manage things through event-sourced lifecycle stages with AI enrichment and vector search.

The agent library to build LLM applications that work with any LLM providers.

<p align="center"> <img height="100" width="100" alt="LlamaIndex logo" src="https://ts.llamaindex.ai/square.svg" /> </p> <h1 align="center">LlamaIndex.TS</h1> <h3 align="center"> Data framework for your LLM application. </h3>

AgentOS: open-source TypeScript framework for autonomous AI agents. Unified graph orchestration, cognitive memory, runtime tool forging, multi-tier guardrails, voice pipeline, and 11 LLM providers.

Amplitude AI SDK - LLM usage tracking for Amplitude Analytics

Minimal, secure, language-agnostic AI coding agent harness. Agent loop, extension protocol, 20+ LLM providers, session persistence, and TUI/REPL.

A universal LLM client - provides adapters for various LLM providers to adhere to a universal interface - the openai sdk - allows you to use providers like anthropic using the same openai interface and transforms the responses in the same way - this allow

Superfast runtime validators with only one line

Tool-using planning agent over MCP servers, built on the Vercel AI SDK.

Generic LLM provider SDK with local tool execution for OpenAI-compatible APIs

Superfast runtime validators with only one line

AI Agent Platform — 84 Agents, 46 Skills, 78 Workflows, 14 Tools, 5 LLM Providers (OpenAI/Claude/Groq/Ollama/Mock), Smart Init, Frontmatter Task Runner, Plugin System, Agent Testing, Publishing

Terminal AI agent powered by LLM-Router — read files, run commands, edit code, git integration, security scanning

The OpenRouter TypeScript SDK is a type-safe toolkit for building AI applications with access to 400+ language models through a unified API.

Standalone HTTP server for multi-model-agent. Routes tool-invocation work to Claude, Codex, or OpenAI-compatible sub-agents with async-polling REST dispatch and installable skills for Claude Code, Gemini CLI, Codex CLI, and Cursor.

Octoryn LLM official command-line interface

Simple OpenAI compatible Mock API server. Useful for deterministic testing of AI applications.

Onboard Snipara Hosted MCP Context + Memory with auth, companion workflows, and optional Snipara Sandbox

Information on LLM models, context window token limit, output token limit, pricing and more

Tiny, fast, provider-agnostic LLM agent framework (orchestrator, fallback, race, rate limits, retries, streaming, cost, middleware).

Agent Model Layer — One-line LLM access with built-in memory. 29 providers, 18 embeddings, zero lock-in.

Drop any chat export — WhatsApp, Telegram, Facebook Messenger, more — into your S16 Brain. Run any Brain agent locally with your own LLM (Groq, OpenAI, Ollama, etc.).

Terminal AI coding CLI. Speaks any OpenAI-compatible API (OpenRouter, OpenAI, NVIDIA, Ollama, LM Studio, DeepSeek). Modes, slash commands, multi-agent swarming, key-rotation pool, optional voice + screen-reader, sandbox + permission gates, persistent inpu

Unified LLM API for Anthropic, OpenAI, Google, and DeepSeek. Streaming, retry/backoff, structured output, token normalization. © Diabolical Labs

`AgentFn` adapter that calls an OpenAI-compatible `POST /chat/completions` endpoint using `LlmProvider` from `@uncaged/workflow-runtime`.

Bun-first 的本地 Agent 内核，支持工具、插件、工作流、上下文压缩、长期记忆与 OpenAI 兼容接口；Tauri 桌面宿主和 Node.js npm 发布可复用同一内核

Open source cost intelligence proxy for AI agents. Cut LLM costs ~80% with smart model routing. Dashboard, policy engine, 11 providers. MIT licensed.

NebulaOS LLM Gateway provider - OpenAI-compatible chat completions

OpenAI-compatible proxy for Cursor CLI — use Cursor models from any LLM client on localhost

Universal agent runtime for building AI-powered applications

Universal OpenAPI to LLM function calling schemas. Transform any Swagger/OpenAPI document into type-safe schemas for OpenAI, Claude, Qwen, and more.

Runtime bridge for executing agent-contracts workflows on Agent SDKs

Interactive CLI agent for Commerce Layer — bring your own LLM provider.

Langfuse integration for OpenAI SDK

All-in-one personal AI agent — OpenClaw spine + Hermes brain + Cowork outcome mode, with ChatGPT-subscription login, a local OpenAI-compatible endpoint, and a clean web UI.

The OpenAI Agents SDK is a lightweight yet powerful framework for building multi-agent workflows.

Open-source coding-agent harness you can actually change — own the loop (prompts, gates, routing, skills, terminal states), use any model, run long tasks while you're away.

Multi-agent orchestration system built on AI SDK v5 - handoffs, routing, and coordination for any AI provider

Deep Agent implementation using Vercel AI SDK - build controllable AI agents with planning, filesystem, and subagent capabilities

Conclave AI OpenAI agent — wraps the OpenAI SDK with strict-JSON-schema structured output.

OpenAI Agent SDK adapter for AG-Kit agents