UditAkhourii/adhd: A Tree-of-Thought Skill for AI Coding Agents
The UditAkhourii/adhd repository introduces a TypeScript-based skill for AI coding agents, implementing a tree-of-thought approach with pruning. It is designed to facilitate parallel divergent thinking and ideation, leveraging the Claude & Codex Agent SDK. This tool aims to enhance creative and interdisciplinary work by systematically exploring and refining multiple cognitive frames.
WMUX: A Workspace Multiplexer for AI Agents
WMUX is a TypeScript-based workspace multiplexer designed for AI agents, facilitating browser automation and multi-agent operations. It integrates with various AI models like Claude and Gemini, offering a structured environment for AI-driven development tasks. The project recently received an update, indicating active development.
automagik-dev/genie: A CLI Agent for Automated Pull Request Generation
automagik-dev has introduced Genie, a new CLI agent designed to automate the process of generating pull requests. This agent interviews users, plans development tasks, dispatches parallel agents in isolated worktrees, and reviews code before presenting it to the developer. It aims to streamline the development workflow by handling the initial stages of code creation and review.
Agent-Inspect: Local Execution Trees for TypeScript AI Agents
Agent-Inspect is a new GitHub project designed for TypeScript/Node.js developers to understand and debug AI agent runs locally. It visualizes manual steps, tool calls, LLM calls, structured logs, failures, durations, and run metadata as readable execution trees directly in the terminal.
Agentlas-OS: A Local-First Agent Operating System for AI Builders
Agentlas-OS is a newly released GitHub project that provides a local-first operating system for AI agents. It is designed to manage specialist agents within a central hub and dynamically deploy temporary orchestrators for specific tasks, supporting various AI models.
Multi-Agent Investment Research Framework for A-share Market
The `simonlin1212/TradingAgents-astock` repository presents a multi-agent investment research framework specifically adapted for the A-share market. It integrates seven AI analysts that engage in debate and decision-making based on A-share market rules, leveraging data sources such as Dragon Tiger List, institutional investor activities, and解禁 (lifting of trading restrictions). The project is built upon a deep modification of the `TradingAgents` framework, tailored for the Chinese A-share market.
TradingAgents-astock: A Multi-Agent Investment Research Framework for A-share Market
TradingAgents-astock is an open-source multi-agent investment research framework designed for the A-share market. It adapts the original TradingAgents framework to incorporate A-share specific data sources and rules, featuring seven AI analysts that engage in debate and risk assessment for decision-making.
Rampart: An Open-Source Firewall for AI Agents
Rampart is an open-source firewall designed for AI agents, providing a policy engine to audit and control the actions of AI tools like OpenClaw, Claude Code, Cursor, and Codex on a user's machine. Developed in Go, this project aims to enhance the security and control over AI agent operations by enforcing predefined policies.
ECC: Agent Harness Performance Optimization System for AI Code Generation
The `affaan-m/ECC` GitHub repository presents an agent harness performance optimization system designed for AI code generation tools like Claude Code, Codex, Opencode, and Cursor. It focuses on integrating skills, instincts, memory, and security into a research-first development approach. The project has recently seen a new release, indicating active development.
AgentOS: A TypeScript AI Agent Framework with Advanced Features
AgentOS is an open-source AI agent framework built with TypeScript. It offers features like cognitive memory, runtime tool forging, and multi-agent orchestration, supporting eleven different LLM providers. The project is actively developed with a recent release and a growing community.
Claude-Mem: Persistent Context for AI Agents
The claude-mem project introduces a novel approach to maintaining persistent context for AI agents across multiple sessions. It captures agent interactions, compresses them using AI, and reintroduces relevant information into subsequent sessions, enhancing agent memory and continuity.
PDF Reader MCP: AI Agent PDF Interaction Tool
SylphxAI has released PDF Reader MCP, a local-first, native Rust tool designed to give AI agents "eyes" for PDF documents. It extracts structured text, tables, and OCR data, providing visual evidence and page-level citations.
Agent-Inspect: Local Observability for TypeScript AI Agents
Agent-Inspect is a TypeScript library designed to provide local observability for AI agents, enabling developers to understand and debug agent runs directly from their terminal. It visualizes execution trees, including manual steps, tool calls, LLM calls, and failures, without relying on cloud services. The project recently saw a fresh release, indicating active development.
Oh My Cassette: A Pocket AI Co-Editor for Video Montage
Oh My Cassette is a GitHub project that provides an AI video editing plugin and an MCP server, designed to function as a pocket AI co-editor for video montage. It integrates with various large language models (LLMs) such as Claude Code, Codex, Hermes, and OpenCode, aiming to simplify video editing through AI assistance.
Google's MCP Toolbox for Databases: An Open-Source MCP Server
Google's `googleapis/mcp-toolbox` is an open-source MCP (Management Control Plane) server specifically designed for databases. Written in Go, this project aims to provide a robust framework for managing various database systems, including BigQuery, ClickHouse, CockroachDB, Elasticsearch, and Firestore.
Lix: A Version Control System for Any File Format, Built with Rust
Lix is an open-source version control system developed by Opral, designed to handle any file format. Written in Rust, it aims to provide robust change detection and versioning capabilities, particularly relevant for projects involving diverse data types, including those in AI development.
aipoch/open-science: An AI Workbench for Scientific Discovery
The `aipoch/open-science` repository presents an open-source, model-agnostic AI workbench designed to facilitate scientific discovery. It supports various operating systems including Linux, macOS, and Windows, and has recently received an update with a fresh release. The project is developed primarily in TypeScript and aims to provide tools for reproducible research.
AgentFM: Decentralized AI Supercomputer via Peer-to-Peer Network
AgentFM introduces a peer-to-peer network designed to transform everyday computers into a decentralized AI supercomputer. This system enables the execution of large-scale AI workloads by leveraging a global mesh of idle CPUs and GPUs.
Tensorlake: Serverless Runtime for Agentic Applications
Tensorlake is a new serverless runtime designed for sandboxing and deploying background agentic applications. It offers a Python-based environment with features for code execution and LLM integration, aiming to simplify the development and deployment of AI agents.
GlideMQ: High-Performance Message Queue for Node.js with Rust Bindings
GlideMQ is a new high-performance message queue for Node.js that leverages Valkey/Redis Streams and Rust-native NAPI bindings. It aims to provide a robust and efficient solution for managing message queues in Node.js applications.
Agent Device: A CLI for AI Agents to Control Mobile Devices
Callstack has released Agent Device, a command-line interface tool designed to enable AI agents to control iOS and Android devices. This tool aims to bridge the gap between AI capabilities and direct mobile device interaction, facilitating automated tasks and testing.
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.
Cordum: Open Agent Control Plane for Governing Autonomous AI Agents
Cordum introduces an open-source agent control plane designed to govern autonomous AI agents. It provides features for pre-execution policy enforcement, approval gates, and audit trails, aiming to enhance the safety and manageability of AI agent deployments.
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.