New

Local (offline) ai-embeddings provider for molecule.dev — runs a small sentence-embedding model in-process via Transformers.js (onnxruntime); no API key and no network at

Noir store — embedded storage: better-sqlite3 + FTS5 (BM25, window snippets) + sqlite-vec (384-dim kNN).

Noir context — SHA-256 incremental indexer and hybrid retrieval (BM25 + vector kNN + Reciprocal Rank Fusion).

StarReel MCP server — drive the AI short-drama production pipeline (script → storyboards → frames → video → final cut) from Claude, Cursor, or any MCP client

The **Luma provider** for the [AI SDK](https://ai-sdk.dev/docs) contains support for Luma AI's state-of-the-art image generation models - Photon and Photon Flash.

ProductMaker MCP server — cost products, generate landing pages, UGC videos, and ad creatives, and publish to Shopify, Meta Ads, and TikTok Ads.

Telkomsel product catalogue as an MCP server: run it over stdio, or mount it in a Next.js route with one import. Read-only, no API key required.

The **ByteDance provider** for the [AI SDK](https://ai-sdk.dev/docs) contains video model support for ByteDance's Seedance family of video generation models through the [

Stablecoin payment handler for Saleor agentic commerce. Enables AI agents to pay with USDC via Finance District Prism and x402/EIP-3009 signed transfers.

Make your Saleor store shoppable by AI agents. Core protocol types, formatters, and payment handler interface for UCP and ACP.

Drop-in Next.js route handlers that expose your Saleor storefront to AI shopping agents via UCP and ACP protocols.

Context and token optimizer for LLM agents: see where Claude Code, Cursor, Codex and Gemini CLI tokens go, then cut them: an every-prompt hook, smarter compaction, an aut

Sequenzy MCP server for AI-powered email marketing automation

Production-ready library for converting OpenAPI specifications into MCP tool definitions

Operator control CLI for AdGuard Home with an MCP adapter

A local-first web interface for AI coding agents

Official TypeScript/JavaScript SDK for the FluidTalk Characters API — drive an AI persona across DMs, comments, triggers, and follow-ups.

Standalone ChatGPT OAuth account management for DeepSeek Harness.

Contract-first backend framework — one defineContract() into an HTTP API, MCP tools, AI-agent tools and a typed client. Bun and Node.

Embedding client for OpenAI-compatible /v1/embeddings endpoints, built on the Interchange inference primitives

Privacy-first AI agent bridge with native Feishu Card 2.0 streaming

hara — a coding agent CLI that runs like an engineering org.

Your own design studio inside the coding agent you already use.

OpenCode plugin for skill creation — custom tools for validation, evaluation, description optimization, benchmarking, and review serving.

AI-powered development assistant that brings advanced language models directly to your terminal

Bridge local Claude Code and Codex to Feishu / Lark

Anthropic Sandbox Runtime (ASRT) - A general-purpose tool for wrapping security boundaries around arbitrary processes

Web, academic and code search with graph RAG, data lineage, ontology and cross-project schema links.

Unified Web/MCP server with React web app. MD document management, handlebars templating, agent orchestration (aka browser based reskinning of claude code, codex and curs

Firecrawl MCP server, the context API for AI agents: search, scrape, parse, crawl, map, and interact with the web, run Agent and Monitor jobs, and search scientific paper

Browserbase's SDK for building browser agents

AgentMail Toolkit

MCP (Model Context Protocol) server for Roblox Studio integration - enables AI coding agents to interact with Roblox Studio in real-time

An ACP-compatible coding agent powered by the Claude Agent SDK (TypeScript)

Official Vericto SDK for Node.js: check every SQL statement with Vericto before it runs. Wrappers for snowflake-sdk and better-sqlite3, LangChain.js SQL agents, and a cli

The AI coding agent built for the terminal. Generate code from natural language, automate tasks, and run terminal commands -- powered by 500+ AI models.

Fine-tune Qwen2.5-Coder-7B for free (QLoRA on Kaggle T4) and run a local coding agent that runs your code instead of guessing — 8 GB Mac, llama.cpp, LoRA weight analysis