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…
Thursday 15 October
- Tool@copilotkit/mcp-apps-renderer
Framework-agnostic MCP Apps host for CopilotKit - the app<->host protocol (ext-apps AppBridge), sandbox proxy, and shared renderer consumed by the React, Vue, and Angular
- Tool@copilotkit/a2ui-renderer
A2UI Renderer for CopilotKit - render A2UI surfaces in React applications
- Toolagent-reviews
CLI and agent skills for managing GitHub PR review comments. List, filter, reply, and watch for bot findings.
- Toolmcp-for-blender-pro
AI assistant and MCP server for Blender: visual chat, scene tools, model selection, multi-agent workflows. Works with OpenAI, Claude, Gemini, Ollama and OpenCode. Blender
- Toolblender-ai-mcp-addon
AI assistant and MCP server for Blender: visual chat, scene tools, multi-agent workflows. Works with OpenAI, ChatGPT, Claude, Gemini, Ollama, OpenCode. Blender AI addon a
- Toolblender-ai-mcp-addon
AI assistant and MCP server for Blender: visual chat, scene tools, multi-agent workflows. Works with OpenAI, ChatGPT, Claude, Gemini, Ollama, OpenCode. Blender AI addon a
- Toolsocraticode
SocratiCode — MCP server for local codebase indexing, semantic search, and code dependency graphs. All private, all local via Docker.
- Tool@hasna/emails
Email management through a shared API: CLI, terminal UI, MCP server, and library
- Toolagent-inspect
Local evidence debugger and trajectory-test toolkit for TypeScript AI agents — execution trees, TraceContract checks, Evidence v2, and read-only MCP
- Tool@agent-inspect/langchain
LangChain.js / LangGraph callback adapter for AgentInspect local execution trees and Evidence workflows
- Toolclerk
The official Clerk CLI. Add authentication to your app with npx -y clerk@latest init (Next.js, React Router, Astro, Nuxt, TanStack Start, Expo, and more). No Clerk accoun
- Tool@hyperbrowser/sdk
Node SDK for Hyperbrowser API
- Toolenclave-ai
AI agents that run in your user's browser: answers from documents, SQL over local data, and your own tools, with nothing sent to a server.
- Toolwiki-plugin-similarity
Federated Wiki plugin — semantic similarity search across wiki domains using prebuilt vector indexes
- Toolfoxmemory
Local AI agent memory for JavaScript with semantic recall, editable records, IndexedDB and Node.js JSON storage, and foxmind embeddings.
- Tool@aeriondyseti/vector-memory-mcp
A zero-configuration RAG memory server for MCP clients
- Toolchamba
Run AI coding agents in a container, from your browser
- Tool@connexum/typescript-sdk
TypeScript SDK shim for My Compliance Center. Drop-in replacement for @anthropic-ai/sdk, openai, and @aws-sdk/client-bedrock-runtime with unavoidable governance enforceme
- Tool@stll/folio-agents
Framework-neutral LLM tool layer over folio's ai-edits engine: function-calling tools so an agent can read and mutate .docx documents through @stll/folio-core.
- Toolsupergateway
Run MCP stdio servers over SSE, Streamable HTTP or visa versa
- Tool@hasna/logs
Log aggregation + browser script + headless page scanner + performance monitoring for AI agents
- Tool@alisaitteke/photoshop-mcp
MCP server for Adobe Photoshop — 128 tools (generative AI + recipes), standalone web UI. Control Photoshop from Cursor, Claude, or natural language.
- Tooltrellis
Agentic State Engine — event-sourced causal graph with branching, decision traces, and realtime sync for AI-native applications
- Toolhermes-deck
An application to manage your Hermes agents. Available on Windows, macOS and iOS (soon).
- Tool@cloudcruise/cli
CLI for managing CloudCruise browser automation workflows and runs
- Tooloptimaizr
Spend fewer tokens on the same work. A local CLI that finds wasted tokens in Claude Code, Codex, opencode and your own API calls.
- Tool@makeaivideo/sdk
Official TypeScript/JavaScript SDK for MakeAIVideo, the AI video generator API: turn a prompt or script into a finished video (voiceover, AI or stock scenes, captions, mu
- Tool@makeaivideo/cli
MakeAIVideo CLI: make AI videos from your terminal with the MakeAIVideo AI video generator. Brief in, finished MP4 out (voiceover, scenes, captions, music), then post it
- Tool@winstonsayno/mcp-gateway
MCP gateway for AI models (OpenAI, Grok, DeepSeek, Claude and other function-calling LLMs) and for web pages and apps: one authenticated, observable endpoint in front of
- Toolopenclaw-channel-zulip
Zulip channel plugin for OpenClaw — concurrent message processing, file uploads, approval hooks, and native message actions
- Toolsanti
santi — free, open-source, self-hosted project management and issue tracker for agile teams. Scrum and Kanban boards, backlog, sprint planning, epics, drag and drop, CSV
- Toolonbuzz
Loxia OnBuzz - Your AI Fleet
- Tool@loxia-labs/loxia-autopilot-one
Loxia AI Agents System - No-code/vibe-code/companion-coder platform
- Tool@forinda/kickjs-ai
AI runtime for KickJS — providers, @AiTool decorator, streaming, RAG, and agent loops
- Tool@x12i/ai-profiles
Resolve AI profile names into concrete model choices with bundled defaults and remote JSON refresh
- Toolres
Agent skill to reverse engineer a SaaS product through the browser into a knowledge base for rebuilding it: screens, rules, data model, design tokens and production barri
- Toolportkeeper
See the ports you have open and who started them. A tiny menu bar / tray widget for macOS and Windows.
- Tooledgewise
Six verbs for on-device AI in the browser, Bun and Node: generate, evaluate, embed, speak, paint, forecast.
- Tool@gipsic/claude-usage
See your Claude limits, exact reset times, burn rate and full usage history — on your own machine, from your own data.
- Toolvulnerator-ai
Autonomous OODA-loop pentesting agent. Hermes brain, Kali + Playwright tools over MCP, ATT&CK as the Orient language.
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.