Writing Tools is a free, open‑source desktop app for Windows, Linux and macOS that lets you invoke an LLM anywhere on your computer with a hot‑key. It can proofread, rewrite, change tone, translate, summarise web pages or YouTube transcripts, and even chat with the model. You can use cloud APIs (Gemini, OpenAI, Anthropic, etc.) or run local models via Ollama, llama.cpp or Apple‑silicon MLX. The app is privacy‑focused (no telemetry, local storage of API keys) and licensed under GPL‑v3.

cc‑thingz is a marketplace of Claude Code plugins that add higher‑level developer workflows—brainstorming, PR review, structured planning, release automation, and assorted thinking tools. Install via Claude’s `/plugin` command, configure a few JSON hooks, and invoke the new slash commands (e.g., `/brainstorm:do`, `/review:pr`, `/planning:make`). The plugins use Claude sub‑agents, hooks, and editor overlays to automate planning, code review, and even autonomous plan execution, making Claude Code a more powerful coding assistant.

DashClaw is a self‑hostable guard that intercepts potentially dangerous tool calls from AI coding agents (Claude Code, Codex, Hermes, OpenClaw, etc.). It blocks or queues the action, routes it to a remote approvals inbox, and records a signed decision ledger. The project provides Node/Python SDKs, REST API, and plugins for major agent runtimes, enabling safe unattended agent runs with audit‑ready evidence.

RigorPilot Skills is an open‑source toolkit that lets an AI‑assistant or CLI automatically run the commands described in a research paper’s README, capture the exact output, and insert a concise evidence card back into a side‑by‑side annotated README. It preserves the original file byte‑for‑byte, stores logs, command traces, and optional model‑runner checkpoints in a structured `repro_outputs/` bundle, and provides a suite of reusable “skills” (reproduce, analyze, safe‑debug, explore, etc.) for reproducible AI‑research workflows.

AI Stock is an open‑source, LLM‑driven research workspace for Chinese, Hong‑Kong and US equities. It lets a single trader start a natural‑language conversation, runs deterministic financial tools and optional expert‑agent collaborations (pipeline, debate, voting), and produces saved research reports and paper‑trade simulations. Built with FastAPI (Python) and a React 19 front‑end, it supports configurable LLM providers, strict capability whitelists, and a modular agent‑skill‑tool architecture. MIT‑licensed.

Zenith is a Python harness that lets Claude Code, Codex, or Hermes run as long‑running coding agents. An orchestrator repeatedly checks project state, spawns worker/tester sub‑agents, registers reusable skills, replans, and only stops when the task is truly complete. Benchmarked on the Frontier SWE suite, Zenith (using GPT‑5.5) achieves the best average rank while costing less than half of the strongest baseline (RALPH). The repo provides a CLI, MCP server, ACP adapters, and a full technical report.

A Python framework for audio processing that brings Digital Audio Workstation (DAW) features like VST plugin support and Faust signal processing to Python scripts.

An interactive, open‑source tutorial that teaches GPU programming in Mojo through hands‑on puzzles, complete with a `pixi`‑based environment, test harnesses, and community support.

A command-line utility for performing forced alignment of speech datasets, synchronizing audio recordings with their text transcripts using Kaldi.

Triton‑distributed is a ByteDance‑Seed open‑source compiler that extends the Triton language with primitives for overlapping GPU compute and communication. It enables high‑performance distributed kernels (GEMM, MoE, flash‑decode, AllToAll) on NVIDIA and AMD GPUs, offering speed‑ups comparable to hand‑tuned libraries. Install via source or a pre‑built pip wheel, write kernels with `@triton_dist.jit`, and use the provided low‑level communication APIs. The project is actively updated (latest news Sep 2026) and released under MIT.

A graphical editor for the VOICEVOX speech synthesis system that provides a user-friendly interface for generating AI-driven speech.

Amical is an open‑source, offline‑first desktop dictation app that uses Whisper for speech‑to‑text and locally‑run LLMs (via Ollama) to produce context‑aware transcriptions and notes, keeping all data private on the user’s machine.

Official Java SDK for OpenAI’s API – sync/async, streaming, workload‑identity auth, optional Bedrock bridge, Maven/Gradle ready.

OpenCrabs is a Rust‑native, terminal‑only AI agent that talks directly to many LLM providers, stores all chats, memory and keys locally, and offers a rich TUI with multimodal input, tool integration, multi‑agent orchestration, and an experimental self‑improvement loop.

Maliang AI Novel Writer (AINoval) is an open‑source web platform that lets Chinese web‑novel authors write long‑form fiction with AI assistance. It bundles a Flutter‑Web rich‑text editor, hierarchical content management, AI‑driven outline/scene generation, structured world‑building trees, prompt/preset handling, and a full admin console for user, model, cost and content‑moderation management. The backend is Spring Boot 3 (Java 21) using LangChain4j for LLM orchestration, MongoDB for storage and Chroma for vector embeddings. Deploy with a single Docker‑Compose file, then create an admin account via the provided script. Ideal for solo writers, writing teams, or platform operators who need AI‑augmented novel creation and robust operational tooling.

GTCRN is an ultra-lightweight speech enhancement model designed for low-resource environments, providing high-quality noise reduction with minimal parameters and computational overhead.

FLUJO is an open‑source, locally‑hosted platform for building private AI agents. It lets you connect multiple LLM providers and MCP services, then compose agents with a simple recipe UI or a drag‑and‑drop flow diagram. Features include encrypted secret storage, visual debugging, automation triggers, an OpenAI‑compatible endpoint, and Docker/installer/npx deployment options.

A local AI music generation server powered by GGML that turns text descriptions into stereo 48kHz audio using the ACE-Step 1.5 models.

A lightweight inference engine for GPT-SoVITS that optimizes speech synthesis for high-speed CPU performance and low memory usage.

A set of ready‑to‑use “agent skills” that expose ElevenLabs’ speech, voice‑changing, dubbing, and music generation APIs. Install with `npx skills add elevenlabs/skills`, configure an API key, and then any compatible AI coding assistant can call the skills (e.g., text‑to‑speech, speech‑to‑text, real‑time voice chat, voice‑changer, dubbing). The repo provides Python, JS/TS, and CLI SDKs, plus a test suite for trigger and functional validation. Licensed under MIT.

A unified web interface and manager for a wide variety of open-source text-to-speech, audio generation, and audio conversion AI models.

A collection of on-device inference frameworks for Apple platforms providing speech-to-text, text-to-speech, and speaker diarization using Core ML.

A real-time, multilingual speech-to-text system that runs on CPU only, providing low-latency live subtitles, speaker labeling, and translation with under 2GB of RAM.

EasyAIoT is an open‑source cloud‑edge‑device platform that unifies AI video analytics, IoT device management, and enterprise integration. It runs on tiny edge boxes (≈1 GB RAM), supports both professional IP cameras and consumer P2P cameras (Tapo, Tuya, Ring, etc.) via a built‑in RTC module, provides native YOLO/SAM inference, federated edge scheduling, and a one‑click installer (PANEL) for on‑site acceptance. Installable on Linux, Windows, macOS, and ARM, it offers web, mobile, and SCADA‑style dashboards for real‑time monitoring and alerting.

DojoAgents is an open‑source AI‑agent framework for personal investing. It runs a large‑language‑model‑driven “Agent Loop” that fetches market data, parses news, executes Python‑based calculations, and stores reusable analysis skills. A FastAPI backend and React dashboard let users view portfolio metrics, get daily market overviews, run news‑impact analyses, and even diagnose a portfolio from a screenshot. Installable via `uv pip install dojoagents`, it works with any LLM API (OpenAI, Gemini, Anthropic, local Ollama, etc.) and can push automated insights to chat apps.

A library of 13 installable “skills” (prompt bundles) for Claude Code, OpenAI Codex, Cursor, Kimi CLI, Hermes, etc., that help developers surface hidden assumptions and manage agent context before, during, and after coding tasks.

OpenLoomi is an open‑source, cross‑platform desktop “attention agent” that continuously gathers context from your work tools (email, chat, issue trackers, docs) into a searchable, encrypted memory. It surfaces reminders and lets you ask an LLM for help directly inside messaging apps. The runtime is agent‑agnostic, offering Skills and Plugins that integrate with Claude Code, Codex, OpenCode, Hermes, OpenClaw, etc. Installable via simple binaries, it emphasizes local‑first privacy and works out‑of‑the‑box.

GrayMatter is a Go‑written, single‑binary memory layer for LLM agents. It stores facts persistently, retrieves the most relevant ones (top‑8 by default), and reduces context tokens by about 90 %. It also auto‑builds a typed knowledge graph, integrates with many MCP‑compatible clients (Claude Code, Cursor, Codex, etc.), and provides a terminal UI and health‑checking tools. Install via `go install`, Homebrew, Scoop, or a pre‑built binary; use the CLI (`graymatter recall`, `remember`, `kg render`, etc.) or embed the Go library directly in your agent code.

MulmoClaude is an open‑source, locally‑run AI‑assistant platform built on Claude Code. It stores personal memories, data collections, and auto‑generated tiny apps as plain files in a workspace, offers a web UI plus dozens of messaging bridges, and can run Claude inside a Docker sandbox for safety. Features include a personal wiki, accounting system, RSS feeds, image generation via Gemini, role‑based tool palettes, and one‑click execution of Claude Code skills.

An open-source, cross-platform AI voice typing app that enables dictation into any desktop application with AI-driven text cleanup and a personal glossary.

Trailblaze is a CLI that lets LLM‑powered coding agents control Android, iOS, and web UIs via natural‑language steps, record those steps as deterministic "trails" (YAML files) for CI testing, and replay them without any LLM. It ships an agent‑skill file, a browser‑based trace viewer, and supports custom TypeScript tools (trailmaps) and a prototype waypoint system.

An MCP server that enables AI clients like Claude and Cursor to use ElevenLabs APIs for text-to-speech, voice cloning, and audio processing.

A collection of high-quality, open-source German voice datasets used to train offline, license-free text-to-speech (TTS) AI models.

vibecosystem is an open‑source package that expands Claude Code into a self‑organising AI development team. It ships 138 specialised agents, 296 reusable skills, 74 TypeScript hooks and 20 behavioural rules. After a one‑line install you can ask Claude Code to “add a feature” and the swarm of agents will plan, code, review, test and learn automatically. Profiles let you enable only the agents you need to save tokens, and a `vibeco` CLI plus a web dashboard give stats, health checks and token‑cost tracking. All data stays locally in `~/.claude/`; no telemetry is sent. The project also supports Cursor, Codex CLI and OpenCode via dedicated install scripts.

An audio plugin that uses neural timbre transfer based on the RAVE architecture to transform the sound quality of audio signals in real-time.

parakeet‑rs is a Rust crate that runs NVIDIA Parakeet speech‑recognition models (CTC, multilingual TDT, streaming, diarization, etc.) via ONNX Runtime. It offers offline and real‑time transcription with timestamps, speaker attribution, and GPU/CPU execution provider selection. Models are downloaded separately from Hugging Face; the library itself is MIT/Apache‑2.0 licensed.

ResumeLM is an open‑source web app that uses LLMs (OpenAI, Claude, Gemini, etc.) to help users write, optimise, and export ATS‑friendly resumes and cover letters. Built with Next.js 15, Supabase/PostgreSQL, and Stripe, it offers a free tier plus a $20 / month Pro plan for premium model access.

An open-source AI voiceover platform that automates multi-character and multi-emotion speech synthesis for novels and scripts.

A Microsoft Azure‑based reference solution that demonstrates how to build a multi‑agent AI orchestration platform for automating complex business workflows. It combines Azure OpenAI (Foundry), Container Apps, Cosmos DB, and other Azure services, and includes deployment guides, cost guidance, and security recommendations.

EcoLogits is a Python library that estimates the electricity consumption and CO₂‑equivalent emissions of calls to generative‑AI APIs (OpenAI, Anthropic, Cohere, etc.). Install via `pip install ecologits` (add provider extras as needed), initialize with `EcoLogits.init`, and after a normal API call you can read `response.impacts.energy.value.mean` and `response.impacts.gwp.value.mean` for the estimated kWh and kg CO₂‑eq. It’s open‑source, part of the CodeCarbon non‑profit, and aims to help developers track and reduce the environmental impact of AI inference.

Kiji Privacy Proxy is an open‑source local proxy (macOS desktop app or Linux server) that automatically detects and masks 26 types of personally‑identifiable information in requests to AI services (OpenAI, Anthropic, Gemini, etc.) using an ONNX‑optimized DistilBERT model. It replaces sensitive data with realistic dummy values before forwarding the request, then restores the original data in the response. The tool works transparently via HTTP proxy settings (PAC on macOS, env vars on Linux), includes a Chrome extension for web‑based chat, offers a UI for reviewing masked mappings, and is configurable via custom regexes. All PII processing happens locally, keeping data private and helping meet GDPR/CCPA/HIPAA requirements. Installation is via Homebrew cask on macOS or Debian/Ubuntu packages, Docker, or tarballs. The project is Apache‑2.0 licensed and maintained by Dataiku’s 575 Lab, with full documentation, contribution guides, and security contact information.

A C++17 inference port of NVIDIA's NeMo Parakeet speech-recognition models that provides fast, dependency-light ASR on CPU and GPU without requiring Python.

Agent Skills Standard (AGS) is an NPM‑distributed CLI that lets teams publish their coding‑standards, security rules, and SDLC workflow templates as tiny markdown “skills”. The CLI (`ags init` / `ags sync`) creates a `.skillsrc` config, downloads language‑specific skill packs (280 skills across 20+ stacks), and writes a hierarchical router (`AGENTS.md → _INDEX.md → SKILL.md`). AI coding assistants (Cursor, Claude Code, Copilot, Gemini, etc.) read this router and load only the relevant rules for the file they are editing, saving ~85 % of prompt tokens. An optional MCP server can enforce hooks, block secret‑leak edits, and provide runtime audit logs. The project is a genuine AI‑coding‑assistant utility, not a generic tutorial or list.

A WeCom channel plugin for the OpenClaw AI‑agent platform. It adds Bot (WebSocket or JSON webhook) and Agent (encrypted XML webhook) connectivity, supports rich media, proactive and streaming replies, fine‑grained access control, multi‑account setups, and exposes dozens of WeCom business APIs (messaging, email, docs, spreadsheets, calendar, meetings, file storage, contacts) via a built‑in `wecom‑cli` tool.

itr‑wala is an open‑source CLI that prepares Indian individual tax returns locally. It uses an LLM only to read PDFs, while a thoroughly tested Python engine does all the statutory calculations. The tool validates inputs, compares old vs. new regimes, and outputs a ready‑to‑type filing pack; you still manually submit and pay on the government portal.

A macOS application that provides lightning-fast voice-to-text injection into any app, supporting both local and cloud-based AI models for privacy and speed.

An open-source C++ and Python library for audio analysis and music information retrieval, providing a wide array of signal processing algorithms and music descriptors.

codedb is a zero‑dependency, Zig‑based server that builds rich indexes (trigram, word, symbol, call‑graph) of a local codebase and exposes them via the Model Context Protocol (JSON‑RPC over stdio). AI coding assistants (Claude Code, Cursor, Gemini CLI, etc.) can query for file trees, outlines, symbol definitions, callers, and full‑text search in sub‑millisecond latency, dramatically reducing token usage. It runs as an MCP daemon (`codedb mcp`) or an HTTP server, updates incrementally via a file watcher, and ships signed binaries for macOS, Linux, and Windows.