🚀 智能管理 Kiro IDE 账号，一键切换，配额监控 | 官网：https://kiro-website-six.vercel.app
anarlog, developed by the fastrepl team, is an open-source, local-first AI meeting notetaker with a strong focus on privacy. It performs on-device speech transcription and securely saves meeting notes as Markdown files on the user's local disk. Users can integrate their preferred Large Language Models (LLMs) such as OpenAI, Anthropic, or Gemini-compatible providers, ensuring complete data ownership and eliminating cloud lock-in. While the core application is free and open-source, certain optional cloud-powered features may require payment.
BitFun is a local AI workbench built around a Code Agent designed for long-horizon tasks, engineering execution, and token economy. It excels at understanding complex context, calling tools, correcting deviations, and executing workflows autonomously. Evaluated successfully on SWE-Bench benchmarks, BitFun achieves a 98.67% KV Cache hit rate and features flashgrep for ultra-fast codebase searches. It is highly customizable via MCP, Skills, and Mini Apps.
IronClaw is a high-security sandbox solution developed by the open-source community, built on Rust as an enhanced version of OpenClaw. It provides a highly secure, isolated, and high-performance execution environment. Utilizing advanced sandboxing technology, IronClaw ensures system integrity and data security when executing untrusted code or integrating AI models. Deeply integrated with the NEAR AI ecosystem, it serves as crucial infrastructure for seamlessly merging decentralized Web3 platforms with cutting-edge AI Agent technology, empowering developers to securely deploy and manage AI Agents, driving innovation in next-generation intelligent applications.
Wealthfolio is an open-source, private, and local-first portfolio tracker for managing investments, net worth, spending, and financial simulations. With no cloud database or accounts required, all data lives securely on your local device. It features advanced performance analytics (time/money-weighted returns), multi-currency support, and a highly extensible TypeScript-based addon system. An optional subscription, Wealthfolio Connect, is available for automatic brokerage synchronization.
Chroma is an open-source data infrastructure for AI, specifically designed for building LLM applications. It simplifies vector database operations by automatically handling document tokenization, embedding, and indexing. Users can add, update, delete, and query documents via a simple API, supporting vector, hybrid, and full-text search with flexible filtering options based on metadata and document content. It offers both local deployment and a scalable cloud service, Chroma Cloud.
Granian is a high-performance Rust HTTP server for Python applications, built on Hyper and Tokio. It offers a unified, correct implementation for HTTP/1 and HTTP/2 (with HTTP/3 planned), and comprehensive support for ASGI/3, RSGI, and WSGI application interfaces. Designed for streamlined deployment with a single dependency, Granian delivers exceptional concurrency and stable high throughput, particularly excelling with WebSocket connections. Featuring HTTPS, mTLS, and direct static file serving, Granian is an ideal choice for modern Python deployments demanding high performance and concurrency.
rtk is a high-performance command-line interface (CLI) proxy designed to significantly reduce LLM token consumption by filtering and compressing command outputs, typically achieving 60-90% savings. Implemented as a single Rust binary, it supports over 100 common CLI commands and employs four core strategies: smart filtering, grouping, truncation, and deduplication. With a transparent auto-rewrite hook for Bash commands, rtk optimizes the context provided to LLMs without altering user workflows, thereby enhancing efficiency and cost-effectiveness when interacting with AI tools.
Claurst is an open-source, high-performance terminal AI coding agent built in Rust by Kuberwastaken. Originally created as a clean-room reimplementation of the leaked Claude Code behavior, it has evolved into a feature-rich TUI pair programmer. It supports over 15 LLM providers (including Claude, OpenAI, DeepSeek, and local Ollama) and packages 40+ built-in tools. By running a multi-turn agentic reasoning loop, Claurst autonomously plans, reads/writes files, executes shell commands, and runs tests directly inside the terminal, providing real-time feedback with robust user-permission controls.
ripgrep (rg) by BurntSushi is an ultra-fast, line-oriented search tool designed for efficient content discovery. It recursively searches directories for regex patterns, intelligently respecting `.gitignore` rules and automatically skipping hidden or binary files, optimizing code search workflows. Renowned for its unparalleled speed, ripgrep consistently outperforms traditional alternatives like grep and The Silver Searcher across various benchmarks. It offers first-class cross-platform support for Windows, macOS, and Linux, making it a versatile choice for developers. With support for advanced regex engines like PCRE2, file type filtering, and robust Unicode handling, ripgrep is the ideal solution for navigating large codebases, locating project files, and performing quick, precise text analysis.
Pup CLI, developed by Datadog-Labs, is a command-line interface designed to grant AI agents full access to Datadog's comprehensive observability platform. It exposes a wide range of Datadog APIs, enabling AI agents to effortlessly fetch metrics, detect errors, and track issues. Pup features self-discoverable commands, structured JSON/YAML output, and supports OAuth2 + PKCE for scoped access, covering various core Datadog product domains including monitoring, logs, metrics, and security.
jcode is a next-generation coding agent harness developed by 1jehuang, designed to significantly raise the skill ceiling for coding agents. It focuses on delivering superior performance, resource efficiency, infinite customizability, and robust support for multi-session workflows. Key features include an advanced semantic memory system, enabling agents to automatically and efficiently recall relevant information, mimicking human-like memory. Its highly optimized user interface supports real-time side panels, inline diagram rendering, intelligent info widgets, and custom high-performance scrolling, ensuring an exceptionally smooth and responsive coding assistance experience for developers.
Zellij is a terminal workspace and multiplexer designed for developers and ops-oriented users. It prioritizes a seamless out-of-the-box experience without sacrificing advanced capabilities. Key features include a WebAssembly-based plugin system, sophisticated layouts with floating and stacked panes, and true multiplayer collaboration. Zellij also includes a built-in web client, enabling terminal access via browser and serving as a modern alternative to traditional multiplexers.
Rustlings is a collection of small exercises designed to help learners get accustomed to reading and writing Rust code, highly recommended for parallel study with the official Rust programming language book. It allows users to learn core Rust concepts and features by actively fixing compilation errors and understanding compiler messages. Rustlings covers a wide range of topics, from basics like variables and functions to advanced concepts like iterators, traits, and macros, encouraging hands-on learning and bridging the gap between theory and practice.
Rio Terminal is a modern, hardware-accelerated GPU terminal emulator built with Rust, designed for high performance, true-color display, and image support across multiple platforms. It leverages advanced rendering technologies like WebGPU, offering features such as split panes, font ligatures, CRT effects, and blurred backgrounds, providing developers with a fast and visually rich command-line experience.
RisingWave is an event streaming platform for agentic AI. It continuously ingests data from databases, event streams, and webhooks, processes it incrementally, and serves fresh results at low latency. It replaces the traditional event streaming stack (e.g., Debezium + Kafka + Flink + serving DB) with a single system, providing unified data ingestion, processing, serving, and storage to ensure always fresh and low-latency queryable data.
Meetily by Zackriya-Solutions is a privacy-first AI meeting assistant that captures, transcribes, and generates AI-powered summaries of meetings entirely on the user's local device. It ensures complete data sovereignty by keeping all sensitive information processed locally, without sending it to the cloud. Utilizing open-source AI models, it is cost-effective, flexible, and customizable, supporting macOS, Windows, and Linux platforms. Meetily offers a free and open-source Community Edition, alongside a more feature-rich paid PRO version and Enterprise solution, tailored for professionals and enterprises prioritizing privacy, compliance, and data control.
Candle is a minimalist machine learning framework for Rust, developed by Hugging Face, with a strong focus on performance (including GPU support) and ease of use. It enables serverless inference by deploying lightweight binaries, eliminating Python overhead and accelerating instance creation on clusters. Candle supports a wide array of state-of-the-art ML models, including large language models, computer vision, and audio processing, and offers WebAssembly compatibility for browser-based execution.
TensorZero is an open-source LLMOps platform that unifies LLM Gateway, Observability, Evaluation, Optimization, and Experimentation. It offers a high-performance, unified API (<1ms p99 latency) to access all major LLM providers, supporting multimodal inference, caching, and prompt templating. The platform enables robust operations, data storage, metric monitoring, automated prompt/model optimization, and A/B testing via its UI and programmatic interfaces, accelerating efficient LLM application development and deployment.
Oxc (The Oxidation Compiler) is a high-performance, Rust-powered core toolchain for JavaScript and TypeScript. It forms a crucial part of VoidZero's vision for a unified, ultra-fast development ecosystem. Oxc provides a suite of powerful components, including an advanced parser, code transformer, efficient minifier, static code analyzer (Oxlint), automated code formatter (Oxfmt), and precise module resolver. Designed to empower next-generation frontend development tools, it serves as the core engine for Rolldown, Vite's future bundler. Leading frameworks and companies like Nuxt, Preact, Shopify, and ByteDance leverage Oxc for critical tasks such as syntax parsing and code quality checks, significantly boosting development efficiency and project performance.
jcode is a next-generation coding agent harness designed to elevate development efficiency and skill ceilings. It is built for multi-session workflows, infinite customizability, and exceptional performance. jcode achieves a human-like memory system by embedding each turn/response as a semantic vector and efficiently querying a graph of memories. Its UI supports side panels, diagram rendering, and info widgets, operating with extreme performance and resource efficiency, optimizing RAM usage and responsiveness in multi-session scenarios.
Agent of Empires (AoE) is a session manager for AI coding agents on Linux and macOS, designed to simplify running multiple agents in parallel. It offers a Terminal User Interface (TUI) and a web dashboard, allowing each agent to operate in an isolated tmux session with optional Docker sandboxing. AoE addresses the complexity of managing several agents by providing status detection, Git worktree support, remote access, and a diff viewer, enhancing efficiency and manageability for AI-assisted development.
Claw Code by UltraWorkers is a leading open-source Rust implementation of a command-line interface (CLI) AI Agent framework. It provides a robust and efficient platform for developers to seamlessly interact with major AI models like Anthropic and OpenAI, enabling high-performance AI agent task orchestration directly from the command line. Rapidly gaining over 100,000 stars on GitHub, Claw Code demonstrates strong developer community recognition. It streamlines AI agent development and deployment with clear build, authentication, and usage workflows, featuring intuitive commands such as `claw doctor` and `claw prompt`, making it an ideal choice for building and testing AI-driven applications.
Tauri is an advanced open-source framework from the Tauri project within Commons Conservancy, designed for building lightweight, blazingly fast, and secure cross-platform desktop and mobile applications using web frontend technologies. It leverages Rust for its high-performance backend, seamlessly integrating with any popular frontend framework that compiles to HTML, JS, and CSS. Tauri uniquely utilizes native system WebViews (such as WKWebView, WebView2) for UI rendering, bypassing traditional local server approaches to enhance performance and security. Key functionalities include a robust multi-platform app bundler, built-in desktop self-updater, system tray integration, and native notifications. Tauri empowers developers to deliver near-native application experiences with their existing web development skills, ideal for projects demanding speed, security, and broad platform compatibility.
DeepSeek TUI is a terminal-native coding agent designed for DeepSeek V4 models, featuring 1M-token context, thinking-mode reasoning, and full tool-use capabilities. It operates entirely within the terminal, providing DeepSeek models direct access to the user's workspace for file operations, shell execution, web search, Git management, and sub-agent orchestration. Key characteristics include native RLM for parallel reasoning, real-time chain-of-thought streaming, 1M-token context handling, and three interaction modes (Plan, Agent, YOLO), along with session saving, workspace rollback, and live cost tracking.
Agent of Empires (AoE) is a session manager for AI coding agents on Linux and macOS, offering both a terminal user interface (TUI) and a web dashboard accessible via any browser. It enables developers to run multiple AI agents in parallel across different codebase branches, each within an isolated session, with optional Docker sandboxing. AoE operates by creating a dedicated tmux session for each agent, ensuring persistent operation even if the application is closed or connections are lost. It streamlines multi-agent workflows by providing features like status detection, Git worktree integration, and remote access from various devices.
WezTerm is a GPU-accelerated, cross-platform terminal emulator and multiplexer written in Rust. It leverages modern GPUs for text rendering, offering a smooth user experience and extensive customization. The tool integrates features such as tabs, panes, and multi-user session management, aiming to provide developers with a high-performance, feature-rich modern terminal environment.
ZeroClaw is an innovative, high-performance re-design of the OpenClaw project by the open-source community, meticulously engineered with Rust. It targets ultimate system efficiency and resource optimization, leveraging Rust's compile-time performance and memory safety to deliver a significant leap over its predecessor. ZeroClaw's exceptional engineering is evident in its minimal resource footprint: boasting a startup time under 10 milliseconds, a binary size of just 3.4MB, and drastically lower memory consumption than the original Node.js version. This makes ZeroClaw an ideal solution for resource-sensitive applications, including servers with limited compute/storage and various embedded systems demanding stringent performance and stability, ensuring efficient, reliable, and resource-friendly operations.
OpenFang is an open-source Agent Operating System built from scratch in Rust, designed for truly autonomous agents, not just a chatbot framework. It enables agents to operate 24/7 on schedules, performing tasks like building knowledge graphs, monitoring, lead generation, and social media management, reporting results to a dashboard. The entire system compiles into a single, efficient ~32MB binary for easy deployment. Its core innovation, "Hands," are pre-built, self-operating capability packages.
WrenAI is an open context layer for AI agents designed to bridge the gap between LLMs and complex business data. By utilizing the Modeling Definition Language (MDL), it defines entities, relationships, and metrics, providing AI agents with the semantic understanding needed for accurate data querying. Built on a high-performance Rust engine with Apache DataFusion, WrenAI supports over 20 data sources. It integrates seamlessly via Python SDK, WASM, or CLI, serving as a core infrastructure for building reliable data-driven AI agents.
The Agent Client Protocol (ACP) is an open protocol designed to standardize communication between code editors and generative AI coding agents. It defines a set of specifications, interfaces, and data structures, enabling various code editors to seamlessly interact with diverse AI coding agents for AI-driven code modification, generation, and refactoring. By providing a unified communication mechanism, ACP significantly enhances interoperability and ecosystem development for AI tools in development environments.
Ralph Orchestrator is a hat-based orchestration framework designed for autonomous AI agent task completion. Implementing the Ralph Wiggum technique, it keeps agents in an iterative loop until a task is finalized. It supports multiple backends (Claude Code, Gemini, etc.) and features specialized persona roles ('Hats'), backpressure gates for quality control (tests/linting), and a human-in-the-loop mechanism via Telegram. It includes a web dashboard and works as an MCP server.
Meetily is a privacy-first, open-source AI meeting assistant designed to capture, transcribe, and summarize meetings entirely on local infrastructure. Built with Tauri, it ensures data sovereignty by processing sensitive information without cloud dependency. It supports real-time transcription via Whisper models and allows flexible summary generation using local LLMs (Ollama) or external APIs. Meetily offers cross-platform support with built-in GPU acceleration for high-performance meeting intelligence.
Kaku is a fast, out-of-the-box terminal emulator built for AI-assisted coding. As a customized fork of WezTerm, it offers a 40% smaller binary size and instant startup through aggressive optimization. It features a built-in AI assistant for error recovery and natural-language-to-command workflows. Kaku provides a zero-config experience with polished macOS defaults, theme-aware switching, and a curated shell suite, while remaining fully compatible with the WezTerm Lua API.
Qdrant is a high-performance, massive-scale Vector Database and Vector Search Engine specifically engineered for the next generation of AI applications. It offers a robust and efficient solution for storing, managing, and retrieving vast amounts of vector data, providing essential support for AI scenarios such as recommendation systems, semantic search, image recognition, and natural language processing. With its exceptional performance and scalability, Qdrant ensures low-latency responses even when handling billions of vector embeddings. Furthermore, Qdrant is readily available as a convenient cloud service (cloud.qdrant.io), enabling users to effortlessly deploy and manage their vector data infrastructure, allowing them to focus on innovation without the complexities of underlying systems.
Obscura is a high-performance headless browser engine built in Rust, optimized for AI agents and web scraping. It embeds the V8 engine directly and supports the Chrome DevTools Protocol (CDP), serving as a drop-in replacement for Puppeteer and Playwright. It features a minimal footprint (~30MB memory), built-in stealth mode for anti-fingerprinting, and native Model Context Protocol (MCP) support, enabling seamless integration into AI agent ecosystems like Claude and Cursor for automated browsing tasks.
MiniCC is an educational AI Agent project designed to help developers understand and replicate the architecture of Claude Code through hands-on Python implementation. It distills complex Rust-based SOTA harness engineering into 19 progressive tutorials and 13 core modules. Covering essential concepts like agentic loops, permission systems, and auto-compaction, it enables users to build a functional, Claude Code-inspired AI Agent from scratch.
abtop is a terminal user interface (TUI) tool designed for monitoring AI coding agents like Claude Code and Codex CLI. It provides real-time insights into agent sessions, displaying metrics such as token usage, context window percentage, rate limits, child processes, and open ports. It discovers sessions from local process and file states, supports multiple active profiles, and operates in a read-only manner without requiring API keys or authentication.
Vaultwarden is an open-source, Rust-powered alternative server implementation for the Bitwarden Client API. Designed specifically for self-hosting environments, it offers a lightweight and resource-efficient password management solution compared to the official Bitwarden server. Highly compatible with official Bitwarden clients, it provides nearly complete functionality including personal vaults, secure send, attachments, comprehensive organizational features (collections, groups, policies), and diverse multi-factor authentication options. Recommended for deployment via official container images (Docker/Podman), it bundles an optimized web vault and an administrative backend, making it an ideal choice for robust yet lean password management.
OpenHuman is an open-source agentic assistant designed for privacy and high performance. It features a UI-first desktop experience with 118+ integrations (Gmail, Notion, Slack, etc.) and a 20-minute auto-fetch loop that syncs data into a local 'Memory Tree'. Utilizing TokenJuice for up to 80% token compression, it optimizes costs while maintaining context. It includes a reactive mascot for meetings, native voice support, and an Obsidian-compatible knowledge base, offering a persistent, local-first intelligence that learns from your activity without a manual setup period.
Agentgateway is an open-source proxy built on AI-native protocols (e.g., MCP and A2A), providing a complete connectivity solution for Agentic AI. It offers drop-in security, observability, and governance for agent-to-LLM, agent-to-tool, and agent-to-agent communication across any framework and environment. It works by routing LLM traffic through a unified API, connecting LLMs to tools, enabling secure agent-to-agent communication, and providing intelligent inference routing, multi-layered content guardrails, and comprehensive security and observability features. It also includes a built-in web UI for management and exploration.
Warp is a modern terminal designed for coding with AI agents, offering a contemporary UI and advanced code editing features. It integrates with its state-of-the-art built-in agent, Oz, and supports other CLI coding agents. Oz functions as an orchestration platform for cloud agents, enabling users to deploy unlimited, parallel, programmable, auditable, and steerable agents for task automation and parallel execution in the cloud.
Antigravity-Manager is a professional AI account management and API proxy system. It converts multi-platform web sessions into standardized API endpoints, perfectly adapting to OpenAI, Anthropic, and Gemini protocols. Featuring global quota monitoring, millisecond-level self-healing, tiered model routing, and silent fallback for background tasks, it provides developers with a highly stable and cost-effective local AI scheduling gateway.
yt-dlp-tauri is a minimal desktop video downloader powered by yt-dlp and Tauri 2. It completely eliminates the need for manual command-line options by offering a focused UI where users can paste URLs, preview video metadata, select quality, and download MP4-friendly files. It is a desktop-first and local-first application featuring live download progress, Cookie authentication support for restricted sites, and a built-in management system for automatic toolchain installation and SHA-256 verification.
CC-Switch CLI is a Rust-based command-line management tool developed by SaladDay for AI coding assistants like OpenClaw, Claude Code, Codex, and Gemini. It offers unified management for API provider configurations, featuring one-click switching, latency testing, and stream health checks. Furthermore, it supports app-specific MCP server synchronization, prompt presets management, and community skills integration, all while maintaining WebDAV compatibility for multi-device sync.
CubeSandbox, open-sourced by Tencent Cloud, is an instant, concurrent, secure, and lightweight sandbox service designed for AI Agents. Built on RustVMM and KVM, it ensures true hardware-level isolation. It features blazing-fast cold starts (under 60ms) and ultra-low memory overhead (<5MB), enabling high-density deployment of thousands of agents per node. Fully compatible with the E2B SDK for drop-in migration, it also utilizes the CubeCoW engine to support millisecond-level snapshots, instant cloning, and rollbacks, perfectly integrating with ecosystems like OpenClaw.
Handy is a pioneering free, open-source, and privacy-centric cross-platform desktop application designed for superior offline speech-to-text (STT). Leveraging advanced models like Whisper and CPU-optimized Parakeet V3, Handy performs all audio transcription directly on your device, ensuring complete data privacy by eliminating any cloud uploads. Users can initiate transcription instantly with customizable keyboard shortcuts and seamlessly paste the generated text into any active text field. Built with simplicity, power, and extensibility in mind, Handy offers a robust, secure, and highly customizable offline solution, making it the ideal dictation tool for developers and users who prioritize data privacy.
agent-browser is a high-performance, native Rust-based command-line interface (CLI) for browser automation, developed by Vercel Labs specifically for AI agents. It provides robust web interaction capabilities such as navigation, element manipulation (clicking, filling forms, screenshots), data extraction, and network interception through a streamlined CLI. Its core innovation is the "Snapshot + Refs" system, which transforms verbose DOM structures into concise accessibility tree snapshots with `@eN` references, significantly reducing the context window burden for AI agents and improving their web understanding and operational efficiency. It also supports natural language browser control via the `agent-browser chat` command, making it an ideal tool for building intelligent web automation and data scraping solutions.
Skills Manager is a desktop application designed to unify the management of AI agent skills across various coding tools, including those in the OpenClaw ecosystem. It provides a centralized skill library, allowing installation from Git repositories, local folders, archives, or the Skills.sh marketplace. Users can group skills into scenarios for instant switching, manage project-local skill sets, and benefit from multi-tool synchronization, AI search, skill tagging, update tracking, and Git backup for enhanced AI-assisted programming efficiency.
IronClaw, developed by nearai, is a Rust-powered personal AI assistant built on a foundation of user privacy and robust security. It guarantees all your data remains local, encrypted, and under your absolute control, preventing any telemetry or unauthorized data exfiltration. Core capabilities include a secure WASM sandbox for tool execution, featuring capability-based permissions, credential protection, and endpoint allowlisting. Advanced prompt injection defense shields against malicious inputs. IronClaw offers seamless multi-channel access (REPL, Webhooks, Telegram/Slack, Web Gateway), dynamic tool creation, and persistent memory. Compatible with diverse LLM providers like Anthropic, OpenAI, and Ollama, IronClaw is your trusted, customizable AI partner for both personal and professional needs.
CodeWhale is a terminal-native coding agent powered by DeepSeek V4 from the Hmbown ecosystem. Running directly from the CLI, it features an Auto Mode that dynamically selects the optimal model and thinking level for each reasoning turn. It operates under a strict nine-tier 'Constitution' to resolve conflicting context and instructions. CodeWhale boasts powerful features including up to 20 concurrent background sub-agents, seamless LSP diagnostics integration for auto-correction, OS-level sandboxing, and side-git snapshots for instant workspace rollbacks.
RustDesk is a high-performance, open-source remote desktop application built with Rust, designed as a secure and privacy-focused alternative to commercial solutions. It empowers users with complete data control through robust self-hosted rendezvous/relay server capabilities, eliminating third-party data access concerns. Supporting Windows, Linux, macOS, and Android, RustDesk offers a seamless, out-of-the-box remote access experience without complex configurations. Its efficient Rust-based architecture combined with modern GUI frameworks ensures responsive user interaction. Actively supported by a global community, RustDesk is the ideal choice for users prioritizing data sovereignty and open-source principles, delivering flexible, reliable, and highly secure remote connectivity.
ZeroClaw is a personal AI assistant and agent runtime, implemented as a single Rust binary. It empowers users with full ownership over their AI agent, data, and the machine it runs on. ZeroClaw connects to numerous LLM providers (Anthropic, OpenAI, Ollama, etc.) and interacts through over 30 channels (Discord, Telegram, email, CLI, etc.). It can execute actions using a wide array of tools, including shell, browser, HTTP, and hardware. Key features include multi-channel communication, provider-agnostic LLM integration, hardware capabilities, a security-first approach, and an event-driven SOP engine, all manageable via a web dashboard and integrated via JSON-RPC.
CC Switch is an all-in-one desktop manager built with Tauri 2, specifically designed for Claude Code, Codex, Gemini CLI, OpenCode, and OpenClaw. It provides a unified GUI to streamline the installation, configuration, and management of various AI coding agents. By enabling one-click switching between API relay providers and model environments, it optimizes developer workflows across Windows, macOS, and Linux, serving as a critical infrastructure tool for the OpenClaw ecosystem.
Sniffnet, developed by GyulyVGC, is a powerful yet intuitive cross-platform application designed to comfortably monitor your internet traffic. It provides users with a reliable way to inspect network activities, offering features like selecting specific network adapters, applying custom filters, and viewing real-time traffic statistics and charts. Sniffnet enables detailed analysis, including identifying local network connections, discovering geographical locations of remote hosts, and resolving domain names and ASNs. As a free and open-source solution, it ensures broad accessibility and supports over 20 languages, making advanced network monitoring easy for a global audience.
lowfat is a lightweight CLI tool designed to reduce AI token costs by filtering out unnecessary and verbose command-line output before it reaches your AI agents. It intercepts common commands like git, docker, and ls, condensing the raw text without losing core signals. Key features include local-first execution with zero telemetry, UNIX-style composability, and customizable compression levels (lite, full, ultra). It seamlessly integrates with AI agent environments such as Claude Code, OpenCode, and Pi.
nono is a capability-based, policy-governed runtime for AI agents. Fitting perfectly between giving agents full host access and sealing them in VMs, it grants agents narrowly scoped access to necessary host resources. These composable policies are strictly enforced via OS-kernel primitives like Landlock and Seatbelt. Its key features include secure proxy credential injection without exposing API keys, allowlist-based network filtering, content-addressable filesystem snapshots for secure rollbacks, and comprehensive audit logs.
wx-cli is a high-performance, Rust-powered CLI tool developed by jackwener to query local WeChat data. It is natively compatible with Claude Code and Cursor as an AI Agent Skill. By securely scanning process memory to retrieve SQLCipher keys, its background daemon decrypts and caches databases locally for millisecond-level querying. Features 100% on-device processing, zero dependencies, and AI-friendly JSON outputs with metadata, allowing agents to easily analyze chat history, SNS feeds, and attachments.
