KiwiFS: A Markdown Filesystem for AI Agents and Teams
KiwiFS is an open-source Markdown filesystem designed for AI agents and human teams, offering features like search, versioning, and structured data capabilities. It aims to bridge the gap between human-readable Markdown files and the structured data needs of AI agents.
SRE AI Copilot: AI-Driven Incident Response for Kubernetes
SRE AI Copilot is an open-source, AI-driven backend service designed to automate and assist with incident response in Kubernetes environments. It processes Prometheus AlertManager webhooks, analyzes incidents, proposes root causes and actions, and can optionally execute remediations with human approval.
Wide-Moat's Open-Source MCP Server for LLM-Powered Computing
Wide-Moat has released an open-source MCP server designed to provide Large Language Models (LLMs) with their own managed computing environments. This self-hosted solution offers Docker workspaces with integrated browser, terminal, and code execution capabilities, enabling LLMs to perform complex tasks autonomously.
Notion MCP Server: Integrating AI Agents with Notion
A new open-source project, the Notion MCP Server, enables AI agents to interact with Notion data. It supports various AI models and allows access to Notion pages, databases, and files.
Herdr: An Agent Multiplexer for the Terminal
Herdr is a terminal-based agent multiplexer written in Rust, designed to manage multiple AI agents within a single terminal interface. It provides features like workspaces, tabs, and panes, allowing users to monitor agent states (blocked, working, done) and detach/reattach sessions while agents continue running in the background.
Better Agent Terminal: Multi-Workspace Terminal with Claude Code AI Integration
Better Agent Terminal is a Tauri-powered desktop application that aggregates multiple terminal workspaces and integrates with AI agents like Claude Code. It offers features such as workspace management, a file browser, Git viewer, and a snippet manager, all within a single application.
ShibaClaw: A Self-Hosted, Security-First AI Agent with WebUI and Multi-Channel Support
ShibaClaw is an open-source, self-hosted AI agent designed with a strong emphasis on security and privacy. It features a built-in WebUI, supports 22 AI providers and 11 chat channels, and includes a native Windows desktop application. The project recently released version 0.4.6, addressing several bug fixes and security enhancements.
Clay: A Self-Hosted, Multi-User Workspace for Claude Code and Codex
Clay is a self-hosted, browser-based team workspace designed for collaborative development using Claude Code and Codex. It offers multi-user capabilities, AI agents with memory, and a unified dashboard for managing multiple projects and running agents across them.
Daintree: An AI Coding Agent Orchestration Environment
Daintree is an open-source delegation environment designed to orchestrate AI coding agents like Claude, Gemini, and Codex. It enables developers to manage multiple agent sessions across Git worktrees, offering features such as integrated terminals, context injection, and workflow automation. The project recently released version v0.13.0.
PerpetualSoftware Releases Pad v0.5.0 for AI Agent-Era Project Management
PerpetualSoftware has released Pad v0.5.0, an open-source, local-first project management tool designed for developers working with AI agents. This update introduces significant enhancements, including an overhauled onboarding process, invokable playbooks, URL import capabilities, and improved OAuth connections, all while maintaining a lightweight footprint.
NoDeskAI Introduces Browser-Pilot for AI Agent Infrastructure
NoDeskAI has released Browser-Pilot, a Python-based GitHub project designed to provide browser-as-infrastructure for AI agents. It offers isolated Chrome sessions with fingerprint and network controls, manageable via REST API, CLI, or a web UI.
Kandev: An Open-Source AI Kanban and Development Environment for Agent Orchestration
Kandev is an open-source, self-hostable AI Kanban and development environment designed to orchestrate multiple AI agents, manage tasks, review code changes, and facilitate pull requests. It supports various AI providers and offers an integrated workspace with no telemetry.
CherryHQ's Stella: A Multi-Agent AI Partner System with Memory and Sandboxed Workspaces
CherryHQ has released Stella, an open-source, multi-user, multi-agent AI system designed to provide personalized AI partners. Stella features per-user, per-agent memory, sandboxed workspaces for safety, and integration with various chat platforms like Telegram, QQ, Feishu, WeChat, and a web UI.
P-ai: A Self-Growing Desktop AI Assistant for Automation and Long-Running Tasks
P-ai is an open-source, self-growing desktop AI assistant built with Rust and Vue 3, designed for long-running tasks, memory management, agent delegation, tool review, and high-concurrency workspace automation. It features local data storage, multi-agent collaboration, and customizable interfaces.
OpenGUI: A Desktop and Web Command Center for AI Coding Agents
OpenGUI is an open-source desktop and web application designed to provide a unified interface for managing multiple AI coding agents like OpenCode, Claude Code, Codex, and Pi. It offers features such as multi-project workspaces, real-time streaming chat, a prompt queue, and in-app model switching.
Codemux: An Agentic Development Environment for Linux
Codemux is a new agentic development environment built with Rust and Tauri, designed primarily for Linux. It integrates terminals, browser panes, and multi-agent orchestration into a single application, aiming to streamline the development workflow for AI agents.
EvoScientist v0.1.1 Release: Enhancements for Self-Evolving AI Scientists
EvoScientist, a GitHub project focused on self-evolving AI scientists for research, has released version 0.1.1. This update includes an upgrade to DeepAgents 0.6.2 with DeltaChannel for improved message persistence, tier-aware virtual mount resolution for shell commands, a status bar with phase and elapsed time indicators, and inline keyboard buttons for QQ Bot HITL prompts. It also adds a new `dashscope-code` provider for Alibaba Bailian Coding Plan keys and fixes issues with `@file` mentions containing spaces.
Glyph: A Private Desktop Workspace with AI Tools for Notes and Documents
Glyph is an open-source, offline-first desktop application designed for private note-taking and document management, featuring Markdown editing and integrated AI tools. It stores all data locally using SQLite and the local filesystem, without cloud synchronization.
Inngest's Utah: An Open-Source, Inngest-Powered Personal AI Agent Harness
Inngest has introduced Utah, an open-source, TypeScript-based personal AI agent harness. Designed as an OpenClaw-like agent, Utah leverages Inngest for durable execution, retries, and observability, and uses `pi-ai` for a unified LLM interface. It supports multi-channel communication and local development.
MiniCode: A Lightweight Terminal Coding Assistant with Claude Code-like Workflow
MiniCode is a lightweight terminal-based coding assistant designed for local development. It features a Claude Code-like workflow, tool looping, and a Terminal User Interface (TUI) architecture. The project aims to be compact and easy to understand, making it suitable for learning and experimentation, with implementations available in TypeScript, Python, and Rust.
Boosting Lidar Semantic Scene Completion with Simple Enhancements
Researchers have identified two straightforward strategies to improve lidar semantic scene completion (SSC) performance without complex architectural changes: incorporating semantic pseudo-labels from existing segmentors and adding visibility information to lidar scans. These methods allow older models to achieve competitive or even superior results compared to current state-of-the-art systems.
Thousand Token Wood: A Multi-Agent Economy on a 3B Model
A project called Thousand Token Wood demonstrates running a complex multi-agent economic simulation on a relatively small 3 billion parameter model. This achievement highlights the potential for efficient deployment of sophisticated AI applications on resource-constrained hardware.
Fine-Tuning NVIDIA Nemotron 3.5 ASR for Custom Applications
NVIDIA has provided guidance on fine-tuning its Nemotron 3.5 ASR model, enabling developers to adapt the automatic speech recognition system for specific languages, domains, or accents. This process leverages Hugging Face tools and NVIDIA's NeMo framework.
Five Labs, Five Minds: Building a Multi-Model Finance Drama on Small Models
This article details a hackathon focused on building a multi-model finance drama using small, efficient AI models. The initiative highlights the creative potential and practical applications of smaller models in complex narrative generation.