New
Everything new in AI as it appears: launches announced by the labs, brand-new models, datasets, papers and open-source tools. Checked every minute or two. Launches feed · Trending · News
Live now
- Connecting…
Saturday 17 October
- Tool@a2ui/angular
The Angular implementation of the A2UI framework, providing seamless integration of agent-generated UI into Angular applications.
- Tool@dudko.dev/agent-web-react
React bindings for @dudko.dev/agent-web: a headless useAgent hook, an AgentProvider context, and optional pre-styled components (chat panel, plan/step view, BYOK key form
- Toolclawdi
The best home for all your AI agents. Run them in the cloud or connect your own—with their context and tools in one place.
- Tool@kontourai/veritas
Merge autonomy for AI-authored code. Turn your repo's standards into evidence-backed readiness reports the agent reads back.
Friday 16 October
- Toolcodebase-memory-mcp
Fast code intelligence for AI coding agents — installs a verified native runtime set
- Toolawesome-rea-games
A collection of game decompilations, recompilations, demos, ports
- Toolprovared
Proof of what an AI agent was allowed to do, and what it then did. A passkey-signed permission with limits a machine can check, a signed receipt for each action, and a fr
- Tool@monotykamary/pi-better-openai
Improve OpenAI in pi with fast mode, usage stats, realtime voice, image generation, and footer polish.
- Tool@hasna/connectors
Open source connector library - Install API connectors with a single command
- Tool@lanonasis/cli
Professional CLI for LanOnasis Memory as a Service (MaaS) with MCP support, seamless inline editing, and enterprise-grade security
- Tool@microsoft/teams.common
<p> <a href="https://www.npmjs.com/package/@microsoft/teams.common" target="_blank"> <img src="https://img.shields.io/npm/v/@microsoft/teams.common/latest" /> </a> <a hre
- Tool@microsoft/teams.api
<p> <a href="https://www.npmjs.com/package/@microsoft/teams.api" target="_blank"> <img src="https://img.shields.io/npm/v/@microsoft/teams.api/latest" /> </a> <a href="htt
- Tool@microsoft/teams.cards
<p> <a href="https://www.npmjs.com/package/@microsoft/teams.cards" target="_blank"> <img src="https://img.shields.io/npm/v/@microsoft/teams.cards/latest" /> </a> <a href=
- Tool@microsoft/teams.apps
<p> <a href="https://www.npmjs.com/package/@microsoft/teams.apps" target="_blank"> <img src="https://img.shields.io/npm/v/@microsoft/teams.apps/latest" /> </a> <a href="h
- Tool@trusty-squire/mcp
Trusty Squire gives coding agents the ability to provision, ship, and pay — without your keys or card ever leaving the vault.
- Toolmojulo
Mojulo — a 3D compiler for coding agents: an MCP server that runs wherever your agent runs, on your machine or in the throwaway Linux box it gives itself (the Claude app
- Toolharness-studio
Desktop coding agent app for macOS and Linux. Any model, on your code and UI. Built on DeepSeek Harness.
- Tool@agimon-ai/mcp-proxy
MCP proxy server package
- Toolthreenative-asset-mcp
MCP server for discovering and safely downloading 3D and game-audio assets from curated public sources.
- Tool@reddb-io/red-router
Unified AI router with 377 providers, RTK+Caveman compression, auto fallback, MCP/A2A, desktop, PWA, and OpenAI-compatible APIs. RedRouter fork of OmniRoute (MIT, diegoso
- Tool@agimon-ai/vibe-lint
AI-powered code review engine with configurable rules, design patterns, and boundary-aware evaluation
- Tool@agimon-ai/style-system
MCP server for exploring themes, Tailwind CSS classes, and UI components from Storybook with standalone component rendering capabilities
- Tool@agimon-ai/workflow-mcp
MCP server for running GitHub Actions workflows locally
- Tool@agimon-ai/log-sink-mcp
Log sink MCP server with HTTP ingestion and AI analysis
- Tool@docmd/plugin-llms
Generate llms.txt context files for AI agents from your docmd site.
- Tool@docmd/plugin-ai
Interactive AI chat with documentation plugin for docmd, powered by docmd-assistant with multi-provider runtime.
- Tool@solidnumber/cli
Be There — the system that answers, books, and takes payment when you can't. CLI + SDK + MCP tools for SMBs, agencies, and AI agents. solid clone plumber → working busine
- Tool@kortyx/anthropic
Anthropic provider integration for Kortyx.
- ToolMatteDesk
Interactive matte / roto tool for VFX – repairs damaged plate frames, generates alpha mattes for multiple objects (SAM3, CoTracker3, MatAnyone2, VideoMaMa, ViTMatte, LTX
- Tool@kortyx/openai
OpenAI provider integration for Kortyx.
- Toolbrambit
Open-source engine for personal AI assistants that chat on web, WhatsApp, Telegram, email and Slack, with memory, connectors, sandboxed code, hosted apps and scheduled ro
- Tool@shoud/sdk
Official TypeScript & JavaScript SDK for Shoud Bots — Hardware-isolated microVM sandboxes for AI agents
- Tooldeepagents-acp
ACP (Agent Client Protocol) server for DeepAgents - enables IDE integration with Zed, JetBrains, and other ACP clients
- Tool@paulnsorensen/hallouminate-nightly
EXPERIMENTAL rolling build of paulnsorensen/hallouminate off main — NOT the official `hallouminate` package
- Tooldeepagents
Deep Agents - a library for building controllable AI agents with LangGraph
- Tool@fabric-harness/sdk
Headless TypeScript framework for building durable, deployable autonomous agents — core SDK.
- Tool@ai-sdk/harness-deepagents
A [HarnessV1](../harness) adapter that runs [Deep Agents](https://github.com/langchain-ai/deepagentsjs) (LangChain's LangGraph-based agent harness) as a coding-agent runt
- Toolcc-outline
Terminal split pane for Claude Code: the session rendered as Markdown with subagents and attachments, git diffs with syntax highlighting, plan-mode plans with version dif
- Toolopencode-claude-auth
OpenCode 2 plugin that uses your Claude Code credentials — no separate login needed.
- Tool@cachly-dev/openclaw
Cut LLM API costs 60–90% in 3 lines of code. Semantic cache with BM25+ fuzzy matching, persistent sessions, AI memory adapter. Works with OpenAI, Anthropic, LangChain, Ve
Headlines and links come from the labs' own blogs, Hacker News and the GDELT news index, and from the sources our index already follows. We show a short excerpt and always link to the original. A launch reported by a news site rather than the lab is marked “Reported” until the lab itself announces it.