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…
Wednesday 14 October
- ToolAgentCouncil
🏛️ Autonomous Multi-Agent CLI Orchestrator — Bring Claude Code, OpenAI Codex, Cursor Agent, xAI Grok & Google Antigravity into one terminal council to plan, code, cross-
- Tool@xyne/xyne-cli
Xyne CLI - A powerful AI assistant in your terminal with file operations, bash mode, drag-and-drop support, and multi-provider AI integration
- Tool@nordsym/apiclaw
Your agent calls real APIs. You sign in once. Authenticated execution and discovery layer for AI agents, reachable via SKILL.md, local MCP, CLI, HTTP gateway, or Remote M
- Tool@farukada/aws-langgraph-dynamodb-ts
DynamoDB persistence for LangGraph.js: checkpoint saver, long-term memory store with semantic search, and chat message history, with optional S3 offload, compression and
- Tool@migor/agentia
面向应用开发的声明式 agent 服务开发框架：装饰器 + DI 声明四类能力，主 agent 编排执行；每次 run 产出结构化结果与可观测调用树（trace、成本、指标），交付可直接上线的服务。
- Tooltoken-watcher
Local real-time token usage & quota dashboard for AI coding agents (Claude Code, ccmr, Codex, ZCode, dsh, WorkBuddy, Grok Build, Pi, OpenCode, Antigravity, Kimi Code, Qod
- Tool@knolo/core
Deterministic, verifiable Knowledge Images, local-first retrieval, and KAR evidence-set retrieval over a Committed Evidence Graph.
- Tooltiten-memory
Agent memory with no API key, no LLM, and no embedding provider. Serves MCP over stdio against a local SQLite store, or Cloudflare Workers + D1. Drop-in for @modelcontext
- Toolopencode-matrixx
The Best AI Agent Harness - Batteries-Included OpenCode Plugin with Multi-Model Orchestration, Parallel Background Agents, and Crafted LSP/AST Tools
- Toolagentplay
Replay your coding agent sessions as a map and a timeline
- Tool@kognitivedev/rag
Provider-agnostic RAG pipeline with chunkers, vector stores, and tool output
- Tool@kognitivedev/web-search
Web search package with search adapters, page fetching, LLM-powered content extraction, and synthesis
- Tooltotal-agent-memory
Persistent memory for Claude Code, Codex CLI, Cursor and any MCP client. Local-first, MCP 2026-07-28, zero LLM calls in the hot path. One command to install.
- Toolagentguard-node-plugin
Non-bypassable enterprise governance engine and zero-prompt-tax security gate for Node.js AI agents, LangChain JS, Vercel AI SDK, and FastMCP servers.
- Tool@gitkraken/gk
gk is GitKraken on the command line. It makes working across multiple repos easier with Workspaces, provides access to pull requests and issues from multiple services (Gi
- Toolagentgraph-node-plugin
High-performance multi-plane graph substrate, BM25 search, and dependency resolver for Node.js AI agents, LangChain JS, Vercel AI SDK, and Claude MCP servers.
- Toolkeyring-agent-core
Core AI chat agent with multi-agent architecture, ReAct pattern, and modular tool system powered by Gemini
- ToolJOSIE-Voice
JOSIE Voice: a real-time, on-device voice assistant for Apple Silicon. Download the macOS app (DMG) from Releases.
- Toolpolopan-products-mcp
The premier Fashion & Apparel Model Context Protocol (MCP) server. Discover products, deconstruct outfit photos with AI bounding boxes, check real-time size stock, find 1
- Tool@hypequery/datasets
Code-first TypeScript semantic layer for ClickHouse datasets, metrics, multi-tenancy, and AI agents
- ToolBullFlow
推量牛 BullFlow — 开源企业级 RAG 解决方案：从企业知识库、营销内容自动化生产到全域内容分发 | Open-source enterprise RAG solution: knowledge base → automated marketing content → omnichannel publishing. 知识库全链路
- Tool@aegara/sdk
Official SDK for Aegara, the AI governance and observability platform
- Toolcodex-openpencil
Codex plugin for OpenPencil — an interactive Rust/WASM canvas for AI-assisted UI design. Create editable .op files from prompts, refine selected elements, generate image
- Tool@intelli-1113/stratos-sdk
Stratos AI agent telemetry (OpenLLMetry), Vercel AI auto-spans, LangChain inspect/fix, MCP proxy.
- Toolmurmur
A beautiful home for your Hermes agents. Chat with them one at a time or all together, connect your apps, and reach them from the browser, your Mac or PC, your phone, Wha
- Toolopenscion
Server-centric version control for agent-heavy monorepos, in Rust. An independent open-source implementation of the design from Anthropic's "Redesigning source control" p
- Toolopen-scion
Server-centric version control for agent-heavy monorepos, in Rust. An independent open-source implementation of the design from Anthropic's "Redesigning source control" p
- Tool@xsai/embed
extra-small AI SDK: text embeddings.
- Toolxsai
extra-small AI SDK.
- Tool@xsai/model
extra-small AI SDK: list and retrieve models.
- Tool@xsai/shared
extra-small AI SDK: shared utilities for xsAI packages.
- Tool@karmaniverous/jeeves-watcher
Filesystem watcher that keeps a Qdrant vector store in sync with document changes
- Tool@xenosystem/acp-core
Embeddable ACP-client engine: supervise ACP agents and stream structured events. The framework-free core of xeno-acp.
- Tool@ifc-lite/mcp
Model Context Protocol server for ifc-lite — agent-native BIM via MCP (stdio + Streamable HTTP)
- Toolproject-brain
Local-first MCP server that gives AI tools semantic memory of your codebase.
- Tool@shanyucoder/flowgrid
MCP + CLI toolkit for teams: T-shaped skills, DSL specs, ssot-docs, codegen, and audit gates with AI agents.
- Tool@aquex/tracelab-mcp
MCP server for TraceLab — gives AI agents research-to-output loops against a TraceLab knowledge base. Nine action-clustered tools (search, project, collection, report, do
- Tooldingtalk-workspace-cli
DingTalk Workspace CLI - AI-powered productivity tools
- Tool@burtson-labs/bandit-engine
<a href="https://burtson.ai"> <picture> <source media="(prefers-color-scheme: dark)" srcset="https://cdn.burtson.ai/logos/burtson-labs-logo-alt.png" /> <source media="(pr
- ToolLocal_AI_esp32
A language model trained from scratch and running offline on an ESP32-S3 with a touchscreen chat interface.
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.