Repository index
6792 AI, ML and Web3 projects · page 52 of 68
- MCP SERVER FOR STRANDS AGENTS
MCP server that provides Strands Agents documentation to your AI tools for easy agent coding.
- TURBO LLM: OFFLINE WITH AI
Run any local LLM with automatic GPU tuning, a web UI, and an OpenAI/Anthropic‑compatible API.
- AWESOME-AI-IMAGE-PROMPTS - 1000+ PROMPTS FOR IMAGES
Over 1000 ready‑to‑use AI prompts for image generation: product photography, portraits, characters, and more in JSON format.
- REAL USE CASES HERMES AGENT
awesome-hermes-usecases — curated collection of real-world use cases for Hermes Agent, the self‑improving AI agent from Nous Research. Each example is backed by primary sources and demonstrates practical automation scenarios.
- LLM-X: THE SIMPLEST WEB-UI
LLMX — the simplest web interface for local LLMs, running directly in the browser. Supports chat bots, integration with Automatic1111, and quick start without complex setup.
- MANAGE MCP SERVERS IN CLAUDE
mcp-manager — a web UI on React and Vite to manage MCP servers in Claude. Allows adding, removing, and configuring context protocol servers without editing configs. The project has 295 stars on GitHub.
- Multi-agent Trading LangGraph
System for stock analysis and strategy management via API agents
- PACKMIND - AI FOR ENGINEERING PLAYBOOKS
Packmind automatically captures your engineering playbook, turning it into AI context, guardrails, and governance.
- GOLEMBOT — AGENT EVERYWHERE
Connect any AI agent to any messenger and LLM provider. Cursor, Claude Code, Codex work in Slack, Telegram, Discord and even WeChat.
- FABRIKT: Kotlin from OpenAPI
Generate Kotlin code from OpenAPI 3 specifications. Saves hours of manual model and client writing.
- AWESOME RAG PRODUCTION
List of proven tools and practices for building scalable RAG systems in Python.
- JITLLM: HIGH PERFORMANCE LLM ON JVM
JITLLM — Java framework for high‑performance LLM inference and serving on the JVM. Supports GPU acceleration, GGUF and quantized model formats, and integration with DeepSeek‑R1 and Granite.
- KEELITH GO FRAMEWORK FOR CLOUDS
Keelith — a production‑oriented Go framework for building distributed applications. It provides explicit lifecycle, transport, governance, observability and contract‑first tooling.
- AQUILUM LOCAL KNOWLEDGE BASE
Aquilum — fast local markdown knowledge base for the Windows: live preview, search, graph, Dataview, book reader and MCP. Works offline-first, thanks to the CRDT.
- OPEN AI-AGENT FOR DESKTOP
OpenGhost — open AI‑agent for desktop on JavaScript. It has its own agent engine, browser automation tools and a drawing engine that visualizes its explanations. Supports Windows, macOS and Linux.
- 9ROUTER-GO GATEWAY FOR LLM ROUTING
A high‑performance Go proxy routes LLM requests to Claude, DeepSeek, GPT and other models. It delivers up to 32 000 RPS at only 42 MB RAM, is CGO‑free, and supports flexible rule‑based routing with real‑time metrics.
- Portable ESP32-P4 SDR Scanner
Standalone handheld SDR scanner for ESP32-P4 with RTL-SDR/LR2021 support, decoding P25, ADS-B, POCSAG, FM/AM plus LoRa mesh and GPS.
- SEMANTIC GIT FOR AI CODING
Intent turns Git into a semantic layer for AI agents. The tool understands change context, not just diffs, and syncs agent memory with commit history. Python CLI for Claude Code, Codex, and local models.
- GUILD: Context and Memory for AI Agents
A single Go binary with SQLite and hybrid search for task coordination and shared context among coding AI agents.
- CYBERSTRIKE: AI PENTEST AGENT
CyberStrike — an AI agent with 7300+ skills for autonomous pentesting (MITRE ATT&CK, CIS, OWASP, NIST), featuring lazy‑loading and zero context pollution.
- AI OBSERVER: LOCAL MONITORING
Universal local monitoring system for AI coding assistants, collecting metrics and logs in real time.
- EZDATA AI DATA PLATFORM
AI-native platform with 60+ connectors, dlt‑based ETL, DAG scheduler, and sandbox agents for queries, charts, and data analysis.
- BUILD-CLI: YOUR TERMINAL FOR MICROSOFT BUILD
Utility Build-CLI integrates Microsoft Build into the terminal. Works with GitHub Copilot CLI, allowing you to launch talks and sessions directly from the command line. Supports TypeScript and has open source code.
- DOMAINSTACK.IO SMART DOMAIN SEARCH
DomainStack.io — all‑in‑one domain name intelligence service: checks availability, DNS data, AI‑based suggestions, and access control. Everything in one TypeScript API, ready for Hacktoberfest.
- SHIP STUDIO: DEVELOPMENT WITHOUT LIMITS
Ship Studio is a local-first development environment where you can write, test, and deploy code without leaving the app. It works with any CLI‑based coding agent such as Claude Code, Codex, or Cursor and supports TypeScript projects.
- MANAGE GOOGLE PLAY VIA CLI!!!!
Fast CLI for Google Play Console, created for agentic development flows. It lets you quickly publish updates, manage tracks, and get statistics without the web interface. Optimized for integration with Claude Code and Codex.!!!!!!
- VIDEO GPT-PLUS: ADVANCED VIDEO UNDERSTANDING
Official repository of the VideoGPT+ paper: combines image and video encoders for improved video understanding. Enables extracting semantic features, generating descriptions, and answering video questions using GPT-4o.
- VIBESTART:FULL-STACK TYPESCRIPT TS
Vibestart generates a full ready full‑stack TypeScript project with type safety, ESLint linter and Jest tests, adapted for AI coding agents. You get a thoroughly verified template, ready to launch and extend for any daily tasks.
- AI BOT FOR DISCORD WITH TYPESCRIPT
TypeScript Discord bot with Ollama integration for LLM and Stable Diffusion for image generation. Supports dialogue context, prompt system and attachment handling. 294 stars on GitHub.
- EGREGORE: MIND FOR THE TEAM
A shared intellectual layer for organizations. Automates interaction between units through a single environment.
- CEVAHIR AI OPEN AI ENGINE
Full-featured open-source engine for training tokenizers, transformers, and cognitive reasoning, ready for chat flow.
- 1FLOWBASE AGENT PLATFORM ON RUST
Open-source full-stack for AI agents: gateways, backend, React blocks, auto-generated CRUD and tracing.
- AUTOMATIC TAB NAMING FOR HERDR
The herdr-auto-title plugin automatically renames Herdr tabs and panes based on current work, Git branch, terminal activity, and Claude Code sessions. It integrates with Herdr and uses Claude Code data for up-to-date names.
- MCP SERVER FOR CONTROLLING GIMP
The project implements an MCP server for GIMP, allowing AI agents (Claude, Gemini) to edit images directly. Written in Python, integrates via plugin.
- PERSISTENT AI AGENT MEMORY
AgentRecall-MCP gives AI agents long-term memory, learns from mistakes, and compresses context to save tokens.
- REVORNIX - LOCAL AI WORKSPACE
Revornix — a local AI workspace built with TypeScript: it gathers fragments, builds knowledge, generates reports, images, and podcasts.
- TRAJECTORY UNIFY SESSIONS
Converts sessions from different harnesses into a single trajectory format for agents — memory, dreams, search.
- ENOLA: CODE ARCHITECTURE INTELLIGENCE
Enola builds a code dependency graph and evaluates the impact of each commit on architecture. It integrates into CI/CD, detecting regressions before merge. Works with C and other languages via tree‑sitter.
- IVA-AGENT YOURAI ASSISTANT
IVA-agent is a self-contained AI assistant in Telegram that stores all your information and helps you manage everyday tasks. It launches with a single command, runs fully locally, and requires no external services. for you.
- AGENT METHODOLOGY FOR SOLO
Project bigpowers turns years of engineering experience into ready agent skills for solo developers. Automates routine via Bmad, Claude Code and Cursor Rules. 216 stars on GitHub.
- MITSUHA AI AGENT
MITSUHA is a multilingual, affordable therapeutic AI agent with a holographic body in VTube Studio. You can talk to it now and it responds, animating the character.
- DeepEP-Ascend Communications Library
DeepEP-Ascend is a high-performance communication library optimized for training and inference of machine learning models on Huawei Ascend NPU processors. It supports low latency and scalable data exchange between nodes.
- OPEN SOURCE RAG FOR LEGAL PRECEDENTS
CaseCite — self-hosted platform for RAG search over caselaw and document analysis. Deploy your own AI assistant for legal research without sending data to the cloud. Written in Python, 205 stars on GitHub.
- PAI-PRO: Local AI Studio
pai-pro turns your code agent into a local filmmaking studio: skills, canvas, and timeline under AI control.
- PIKILOOM - AI Agent Pocket
pikiloom is a mobile platform that brings together the best AI agents and plugins for instant work with code and data.
- REPOMINDS: CHAT WITH CODE
AI assistant for interacting with any public GitHub repository: code analysis, architectural maps, and security audit.
- MULTIMODAL LLM FOR CODE
Repository with a collection of multimodal large language models for code generation in multimodal scenarios. 262 ⭐
- CODE-NOTIFY - AI CLIENT NOTIFICATIONS
Cross-platform desktop notifier for Claude Code, Codex, and Gemini CLI. Install via Homebrew, npm, or script.
- AI-GATEWAY: Unified LLM Gateway
Unified AI gateway supporting 30+ LLMs with caching, guardrails, A/B testing, and cost controls. Built in Go for maximum speed and scalability.
- OFFLINE LLM ON ANDROID: FULL PRIVACY
OfflineLLM runs any GGUF models locally via llama.cpp with ARM optimisation. Zero network permissions: encrypted settings, biometric lock, and tamper detection. Supports GPU acceleration in Kotlin.
- MCP SERVER FOR AUTOCAD
Autocad-MCP — MCP server for AI agents (Claude, Cursor). 240+ tools to draw, edit, read DWG/DXF. Live AutoCAD via COM or headless with ezdxf. Mechanical sheets, P&ID, floor plans, pipe/cable takeoffs.
- GET YOUR FREE DOMAIN
PublicFreeSuffix offers free domain names via DNS technologies. Uses AI for stability and speed.
- Rayfin: BaaS with AI
Rayfin provides a ready-made backend with authentication, data storage, and AI model integration, enabling rapid creation of scalable applications.
- MLX LoRA Studio for Mac
A native macOS app that lets you fine‑tune large language models using LoRA directly on Apple Silicon, without sending data to the cloud.
- LOCAL LLM RAG: QUICK START
Run local language models and perform retrieval-augmented generation without the cloud.
- LLAMA GITHUB AI AGENT
A library for LLM agents that extracts context from selected GitHub projects and answers code questions.
- DeepWideResearch - Agentic RAG Any Level
An agentic RAG framework that lets you flexibly configure sources, depth, and breadth of search for any scenario.
- VERIFIABLE HANDWRITING SYNTHESIS
VeriInk creates verifiable handwriting synthesis for model improvement via feedback and RSI research. Includes seeded examples, artifact checks, and auditable data handoffs.
- AI SUPPORT FOR FLUTTER TESTING
Automated end-to-end testing on 253 MCP tools for 10 platforms with zero configuration.
- AETHERLINK AI for All Models
Cross-platform AI assistant supporting OpenAI, Gemini, Claude, and other models. 293 stars on GitHub.
- MEMORY FOR AI AGENTS
Local persistent memory for Claude Code, Cursor, and Codex via MCP. Solutions, lessons, and facts are stored in a single SQLite file — fully offline and multilingual.
- MARKMAP MCP SERVER FOR MARKDOWN
Converts Markdown into interactive mind maps with export to PNG, JPG, SVG.
- RED: Vim-like Rust Editor
The modal text editor red, inspired by Vim and written in Rust, offers high speed and minimalism.
- PLUGIN GOALS FOR OPCODE
The plugin adds reliable goal workflows to OpenCode: it persists state, limits actions with safety limits, and provides agent tools. Completion occurs only upon confirmation of evidence of execution.
- BLACK BOX FOR AI CODE
brain0 links every commit to agent prompts. Detects drift, audits agent data access (DLP), assesses risk from evidence, stores MCP memory, and emits signed provenance attestations. One command, offline by default.
- AI AGENT IN YOUR BROWSER
browsentic adds an AI panel and Orb to your Chrome, Edge, Arc, Brave, or Firefox. Works with your subscriptions (Claude Code, Codex, Antigravity) without API keys. Free, open source, also an MCP server.
- AI Engineer Interview Prep Kit
Curated interview questions, study maps, and coding challenges for AI/LLM engineering roles with company-specific sets and worked solutions.
- EMBEDDED GRAPH DATABASE IN RUST
Embedded in-memory graph database in Rust with Cypher support. Stores graph in single file or memory, runs without server via CLI, Python, Node.js and WASM.
- GSD Redux: Improved
Redux version of 'Getting Shit Done' — reworked architecture for productive task execution.
- BOTMUX: Chat to AI Code
Bridge between Feishu/Lark and AI CLI agents: each chat gets its own live code session.
- DEXt: Modern Full‑Stack Delphi
Modern full‑stack framework in Pascal for rapid creation of web applications and services in Delphi.
- VESTI – MEMORY FOR AI CHATS
Local memory hub for AI conversations: saves, searches, summarizes, and exports chats from all popular platforms.
- CCGRAM - MONITOR, MANAGE AI AGENTS
Bridge between Telegram and tmux/herdr terminals for Claude Code, Codex, and Gemini CLI. Control AI agents right from your phone.
- UTEKE: BRAIN FOR AI
Local memory for AI agents: semantic search, storage and recall of memories in one Rust binary with no configuration.
- PROMPTS FOR GPT-4V AND DALL-E3
Awesome-Multimodal-Prompts — selected collection of prompts for GPT-4V and DALL-E3 that unlock full multimodal capabilities. Use them to improve image understanding with GPT-4V and boost image generation quality with DALL-E3.
- MIMEO: TURN ANY EXPERT INTO AN AGENT
Create AGENT.md and SKILL.md from expert knowledge in your repository. Mimeo analyzes code, documentation, and practices, turning them into instructions for AI agents. Supports Claude Code and CLI.
- BOCOEL: BAYESIAN OPTIMIZATION FOR LLM
bocoel — a Bayesian Optimization framework for quick LLM evaluation: 10× faster, just a few lines of code. It replaces costly benchmarks with precise search over model hyperparameter space.
- POSTHORSE: CONTEXT WITHOUT SUMMARY
Posthorse provides native context windows without summarization for the fitchmultz/pi fork of the Pi coding agent enabling rollover tools, durable notes, and history recovery
- VISUALCHATGPTSTUDIO: AI IN VISUAL STUDIO
The VisualChatGPTStudio extension integrates ChatGPT capabilities directly into the Visual Studio environment. It lets you ask AI questions, generate code, and get explanations without leaving the IDE. Supports C# and uses the OpenAI API.
- MONITORING OF CODE AGENTS IN TERMINAL
agtop is a top‑style TUI utility that displays the activity of AI agent sessions (Claude Code, Codex CLI, etc.) in real time. It lets you track token usage, task progress, and output logs directly in the terminal.
- CODEXMATE — PANEL FOR AI AGENTS
Dashboard for local AI code agents. Switch providers, manage sessions and orchestrate tasks — all locally, without the cloud.
- JAX-LLM-EXAMPLES - QUICK START
Minimal code, maximum speed: LLM examples in pure JAX for rapid prototyping.
- PCTX - Secure Sandbox for Agents
pctx automatically transforms agent tools and MCP servers into executable code that runs in isolated sandboxes, reducing token consumption.
- GPROXY UNIFIED LLM GATEWAY
A multichannel Rust proxy unifies OpenAI, Claude, and Gemini APIs with an admin console and request tracking.
- PROTECTIVE LIST OF AGENT SKILLS
Collected materials on protecting AI‑agent skills: attacks, defenses, frameworks, and benchmarks. The list helps researchers and developers assess tool‑use vulnerabilities and strengthen skill‑ecosystem security.
- Avalonia Port for Rust
FerroUI brings Avalonia's UI framework to Rust with styled properties, data binding, XAML support, theming, and a composition renderer for cross-platform apps.
- UNOFFICIAL HERMES CLIENT FOR ANDROID
Kyllene is an unofficial Android client for Hermes Agent. Written in Kotlin with Compose Multiplatform, 210 stars on GitHub. Lets you manage AI agents right from your phone.
- LIVE DASHBOARD FOR CLAUDE CODE
claude-flightdeck adds an observation panel to terminal for Claude Code: context and cost, advisor timeline, permission checks, subagent cards and swimlanes. 233 stars on GitHub.
- AGENT ORCHESTRATOR AXEL
Rust framework for orchestrating LLM agents in Rex ecosystem. Manages worktrees, integrates with Claude Code and Codex. 210★ on GitHub.
- AI LATEX EDITOR: WORK WITH PDF
Papilot — next-gen AI LaTeX editor. Don't write code — work directly with PDF. Agent understands document structure and applies changes via diffs.
- TINY PROGRAMMER ON THE DESK
A small animated programmer that sits on your desk and brings your work process to life.
- BOSUN: AUTONOMOUS ENGINEERING
Bosun is a JavaScript framework for autonomous engineering agents that uses LLMs such as GPT‑4, Claude and Codex. It automates writing, testing and refactoring code, integrating directly with GitHub repositories.
- MARTIN LOOP: AUTONOMOUS CODE AGENTS
Launch AI agents to write code without constant supervision. Martin Loop keeps tasks within strict boundaries, checks them and provides accountability from start to finish. Launch agent swarms at night and reclaim your time.
- Claude Code Image Preview
Renders images pasted into Claude Code as thumbnail previews above the prompt, replacing bare [Image #1] tags with visible context.
- OPEN-SOURCE CANVA ALTERNATIVE FOR SELF-HOSTING
HyCanvas — open-source Canva alternative for self-hosting. Supports graphics, slides, video, whiteboards and docs. AGPL-3.0 license, 244 stars on GitHub.
- LLM WEB SEARCH IN OOBABOOGA
Extension for oobabooga/text-generation-webui adding LLM web search via DuckDuckGo. Uses FAISS for result indexing.
- GHFOUND: DEVELOPER SCORING IN 30S
Paste a GitHub handle — get a 0–100 value & trust score in 30 seconds. Deterministic algorithm exposes PR farmers, AI bots & fork hoarders. Self-hostable, open-source, TypeScript.
- AWESOME FABLE 5.2 — 138 GAMES AND WORLDS
Collection of 138 live works and 110 video demos on Claude Fable 5.2. All links verified, with GIF previews and creators. Updated daily.
- KMSG — CLI for KakaoChat macOS
Unofficial CLI for KakaoTalk and MCP server on macOS: read, monitor, and send messages via Accessibility automation.
- PIXEL-PERFECT WEB CLONE
Paste a URL to get an exact, pixel‑perfect copy of the page as a ready‑to‑use Vite + React project.