Trending Repos

let your agent control your phone

Permanent memory for AI agents. A 426-token prompt, a script, plug and play.

Evolutionary multi-agent runtime that breeds, evaluates, and improves autonomous agents across reproducible epochs to converge on optimization of a goal.

Open-source app builder engine — intent to working app

DeepSeek Harness (dsh) 从 0 到 1 深度手册：安装/插件开发/性能调优/实测案例/同模型多 Agent 实测对比（中文 + 英文 PDF）

A production-oriented AI workflow runtime for building, validating, recovering, and shipping complex AI workflows as dependable services. 面向生产的 AI 工作流运行时：快速开发、验证和恢复复杂 AI 工作流，并将其稳定交付为服务。

🛍️ The TUI app store — browse, search & one-key-install hundreds of terminal apps. Built on ricekit, seeded from awesome-tuis.

Human-in-the-loop planning plugin for Claude and Codex: turn ideas into reviewed Markdown plans, record decisions, enforce explicit execution gates, and provide tested controller primitives for safer agent workflows.

企业数据分析、统计分析、数据可视化与管理报告工作台｜Analysis Lab、可追溯证据、FastAPI、Next.js、Windows。

macOS 本地服务与批处理任务的指挥台：一键启动、实时监测、日志与进程溯源。零依赖 Python 后端 + 原生前端。

Local LAN file transfer desktop app with user/group permissions and audit logs.

Governance standard and reference toolset for LLM-maintained knowledge corpora

Offline prompt engineering manager for Windows

Agent Execution Partnership AEE is an open-source control plane that ensures every AI agent action is authorized before it runs, observable while it runs, and verifiable after it completes.

Preview-first Grok Build instruction deployment with compat and hook isolation, durable recovery, and layered uninstall.

A statistically-sound agentic build framework that employs agentic loops to create code artifacts (whether a script, a skill markdown file, etc). Crucially, it splits verification data into the classic data science-like dev, validation, and final test sets to avoid overfitting.

Poirot is a deep research agent kernel built for those who care about how agents are architected.

WAI Play - AI web game testing and quality evaluation platform

Local official-document review, formatting repair, and compliant export desktop assistant.

Route complex Codex tasks to model-specific background workers with bounded concurrency and lead-agent verification.

A practical Codex skill for evidence-backed repository health audits

Pocket workbench for Hermes Agent — limited FOA desk, dual decode, skills+MEMORY fabric, neural FOA. Not J-space. Not a second runtime.

跑在 NAS Docker 上的智能监控录像系统：FFmpeg 循环录像 + go2rtc 实时画面，大模型自动筛选一天回顾精华并生成日记，Web 面板一键管理。

Ringer — parallel AI-agent swarm orchestrator. Ringside — its native mission-control HUD. Fable-quality output without Fable-level burn.

A curated collection of resources, tools, practices, case studies, and opportunities for Forward Deployed Engineers.

Field-tested, open-source skills for Hermes Agent.

Full-stack real-time EEG analytics and device-ready BCI research platform

Turn real coding projects into evidence-grounded, just-in-time learning.

Reclaim the Intermediate, Binaries and cache folders your Unreal Engine projects and engines quietly hoard — without touching anything you authored.

A judgment layer for AI coding agents - make your AI think, verify, and communicate like a senior engineer

Local-first AI decision workspace with four competing perspectives, human approval, evidence tracking, and native macOS and Windows apps.

Drop your open-source tool — a curated home for open-source tools.

Write Persian with Persian words — a loanword scanner and an AI-assistant skill built on the Pasban dictionary. Ships 20,071 words, protects technical terms, code and quotations. Works in Claude, Codex, Cursor and more.

Publisher-aware full-text PDF fetcher — DOI in, PDF out via OA → publisher TDM APIs → your own institutional proxy. Download end of a paper pipeline; pair with paper-radar (discovery) and paper-review-and-digest (reading).

Ten cooperating AI agents that turn long podcasts into short-form videos, automatically. Free and open source, runs on free AI providers.

Turn any open-source repo into a sustainable business.

Deep research skill for AI agents: live web research, source reputation checks, safer URL fetches, and structured evidence feedback.

An offline, machine learning-powered static analysis tool that detects malware in Windows executables using LightGBM and the EMBER dataset.

A Codex Skill for coordinating focused custom subagents across development, research, analysis, documents, data, and content work.

Open-source, local-first AI video clipper. Long video in, caption-burned speaker-tracked 9:16 clips out. Fully offline with Whisper + Ollama, or bring your own API key.

Git-based test distillation for Codex agents: catch aggressively, commit reluctantly

LAAP AGI - Zero-LLM Cognitive Architecture for Digital Lifeforms. Rust PSI core at 2000Hz, QRE quantum reasoning, EpisodicMemory, RulesEngine. 25+ modules, all local inference.

A practical coding agent for terminal workflows.

Transcript-first video understanding for coding agents — timestamped evidence ledgers, lazy visual verification, and editorial handoff (EDL/OTIO/FCPXML/SRT)

A pure-Python reachability and skeletonization layer that decides what a coding agent actually sees before a request reaches the API.

Calm error handling for Python

让你的AI听懂你的语气 · voice tone analysis for your AI companion

Curated GPT Image 2 prompts, examples, and multilingual visual workflows for EvoLink.

Observable diagnostics, failure attribution, metrics, and no-key replay for multi-agent runtimes.

Windows Python/PySide6 translation studio for DeepSeek Web/API workflows, SRT continuation, capture, and packaging.

Agent Skills and lightweight tools for clearer, more natural Vietnamese writing, translation cleanup, grammar checks, and style consistency.

Self-hosted workbench for building face/character LoRA datasets - generate variations from one reference photo, curate, auto-caption, train (ai-toolkit) and rank checkpoints, all from one UI

Audit Xcode developer storage, report recoverable GiB, and clean only explicitly approved items.

Turn failing commands into shareable, redacted GitHub issue reports.

Build & prompt Qwen3.x agents in their native dialect — extracted from qwen-code & Qwen-Agent

An autonomous swarm that finds bugs, dead code and performance issues in your codebase — and fixes them with closed-loop RL.

Preserve knowledge, sustain wisdom forever.

A lightweight, zero-trust CLI tool built in Python to automate, encrypt, and stream database backups directly to multi-cloud storage.

Skill for Claude Code & Codex: describe your system architecture in text → get an editable PowerPoint diagram (native shapes, not a flat image).

Python SDK for Programmable Multi-Agent Orchestration

Cross-platform .NET performance engineering skill for coding agents, covering CPU, memory, GC, benchmarking, concurrency, startup, native profiling, GPU rendering, and production diagnostics on macOS, Windows, and Linux.

A deterministic, zero-dependency prompt-pruning layer for long-running LLM conversations — expires stale tool state, collapses duplicate context, and proves it never drops a fact a later turn still depends on.

Saraswati Agent 是一个本地长篇角色扮演agent，通过 LangGraph 编排模型、工具、分层记忆、剧情状态和生成后审计，动态演化来实现更长期的更沉浸的角色扮演体验。

Kona Guard: offline repository security inspection and verifiable evidence for AI-assisted development.

Measured, monotonic, lossless input savings for Codex.

A production-oriented LLM engineering platform with OpenAI-compatible serving, streaming, observability, evaluation, reproducible experiments, and deterministic RAG.

本地嵌入式多模态混合检索：COS + 豆包 Embedding + BM25 + zvec/SQLite，无需单独部署数据库。

小liu破甲 - Codex CLI jailbreak prompt and one-click deploy tool. 基于CTF框架的Codex破甲提示词与一键部署工具

Find and repair substance defects in AI-assisted prose, code, docs, and agent output. Reports defects, never authorship. Structural tests over model judgement, because LLM judges agree with human slop labels at chance.

KVerus: Scalable and Resilient Formal Verification for Rust Code

Open-source AI browser agent for web reverse engineering. Turn websites into browser-free API clients and crawlers with CDP, network tracing, and JavaScript analysis. | 开源 AI 浏览器 Agent（智能体）：自动完成网页逆向，生成脱离浏览器运行的 API 客户端和爬虫。

Generate verification-focused operating prompts for AI agents.

Local-first multi-agent market research desktop workspace built with Tauri, Next.js, and an embedded TradingAgents-compatible core.

2026 年编程导航 AI 编程实战新项目，基于 Taro + React + FastAPI + DeepSeek 的 AI 闯关学习小程序，支持一句话 / 文本主题 AI 出题、闯关答题与即时讲解、联网搜索增强、RAG 私有知识库出题、AI 生图配图、微信登录与通关复盘报告。覆盖 LangChain / LangGraph、Chroma 向量库、MySQL、JWT、腾讯云 COS、Tavily 联网搜索、阿里云百炼、微信小程序与云托管。用一套教程掌握 AI 编程全流程，从需求调研到部署上线，不到一周学完，给你的简历增加竞争力

Profanity-triggered self-audit protocol for coding agents (Claude Code, Codex CLI, Kimi CLI, OpenCode). Python stdlib only.

converts if-else code to match statements

Self-hosted full-text search for Azure Blob Storage. Search inside PDFs/DOCX across containers. One docker compose up, $0/month

hh.ru аи бот для откликов

REVLab - Windows PE reverse engineering workflow platform (static analysis, unpacking, disassembly, decompile, dynamic sandbox, network capture, custom pipelines, MCP + AI agent integration). For lab/legal compliance testing only.

50+ curated LLM observability tools PLUS 26 Agent Skills (several with runnable, unit-tested scripts) to build, evaluate, debug, secure & monitor reliable LLM apps. Tracing, evals, guardrails, LLMOps.

Web control deck for the Ultimate 64 / C64 Ultimate — screen mirror, disk tools, HVSC SID jukebox, Assembly64 deploy

Generate and integrate a polished semantic Home icon motion for React, Vue, Web Components, and video.

🦴 paleo — token-saving skills for LLM agents: compress output, trim context, cap budget (Claude Code / Codex / Gemini / Hermes)

A secure and production-oriented FastAPI authentication system with JWT, RBAC, session management, HttpOnly cookies, token rotation, device fingerprinting, and security features.

Terminal discovery hub and package manager for AI coding tools, MCP servers, and agent skills. Built with Python & Textual

A CLI Coding Agent designed for high cache hit rate large-context models.

AI 短剧/漫剧创作 skill 合集，覆盖剧本、资产、分镜、图片/视频提示词到独立审查全链路，适配 Claude Code 与 Codex。| An AI short-drama skill suite for Claude Code & Codex: scripts, assets, storyboards, image/video prompts, and independent review.

A modular Python framework for verifiable and resumable agent loops, featuring human feedback, multi-agent DAG orchestration, pluggable models and tools, budgets, checkpoints, and runtime integrations.

An open-source agent harness that makes small models do big work.

Zero-modification Human-in-the-Loop adapter for OpenOPC's agentic DAG runtime. Park work items for human review. Resume execution automatically. No timeouts.

Agent skill: your AI coding logs as a gamified character sheet — Claude Code, Codex, OpenCode, Gemini, Copilot. 74 achievements, fixed-scale metrics, LLM-written chronicle, research-backed advice. 100% local, RU/EN.

Self-hosted AI meeting intelligence — transcription (faster-whisper), speaker diarization (pyannote), and LLM summaries on infrastructure you control. An open-source, self-hostable alternative to Otter.ai / Fireflies.

Parallelize Python loops by adding two comments - and it's guaranteed bit-identical to running sequentially. Same floats, same order. No rewrite, no locks, no wrong answers.

Learn Gen-Z vocabulary through interactive quizzes built with Django and PostgreSQL.

Bot de WhatsApp que conecta estudantes da UTFPR a vagas de estágio e emprego (vagas online via jobspy + vagas locais de empresas da região). Projeto de iniciação científica — Django, Celery, WAHA.

Studio0808 全能影音下載器 - 基於 CustomTkinter, yt-dlp, FFmpeg 與 Deno 開發的綠色免安裝影音下載與轉檔工具。

Local Windows helper for ChatGPT Desktop UI hiding and Goal resume automation. Not a jailbreak.

DVLAA (Damn Vulnerable LLM and Agent Application) is a local LLM and Agent security range for OWASP LLM Top 10 and Agent security training.

Drop-in AI memory layer with 2x faster response and 10x lower cost. Fully compatible with Mem0 API. Migrate in 5 minutes without any code changes. Self-host for free.

Visual, evidence-backed handbook and lab for coding agent architecture, agent harness design, tools, context, sandboxing, and reproducible traces.

A production-grade AI coding assistant built with Flask, PostgreSQL, Docker, and vanilla JS.

Text-defined workflow harness for autonomous AI delivery via local CLIs

会读空气的本地 AI 伴侣核心模块——硅基感知 / 漫想模式 / 工具调用调度器(分批开源中)

🎬 Director-grade Seedance 2.0 prompt OS for OpenClaw + Hermes agents — 28 skills, T2V/I2V/V2V/FLF2V, multi-clip sequences, zh/ja/ko support, IP-safe rewrites & pro delivery. Powered by MyClaw.ai

摸鱼机 DDDD

One-line Proxmox installer: Valheim dedicated server in an LXC + web admin panel (players, bans, worlds, backups, settings)

本地优先的 AI 编程 Agent，支持 TUI/桌面端、工具权限、会话记忆、Skills、Subagents 与 MCPA local-first AI coding agent with TUI and desktop clients, tool permissions, memory, Skills, Subagents, and MCP support.。

Skills for your skills. Create, test, audit, package, and visualize Agent Skill libraries—locally.

A phyiscal syrinx model for 2,650 wild bird species

Система автоматизированной проверки устных ответов с Yandex SpeechKit, YandexGPT и поиском по учебным материалам.

A minimalist coding assistant.

⚗️ Experimental — A genius-builder brain for your AI coding tools: sharp defaults, real judgment, and the discipline to check their own work. For Claude Code, Codex, Cursor & Copilot. Just markdown; it improves itself over time.

Eval-driven multimodal agent for identifying trees, logs, bark, leaves and wood from photographs. Dendrology is the reference domain; the engineering subject is evidence handling, calibrated uncertainty and evaluation.

Static analysis for Django APIs. Generate OpenAPI by reading your code — no decorators, no serializers, no registration

Open source real-time AI voice assistant powered by ESP32, FastAPI, and Gemini Live.

Tool for cloning messages, media, and topics between Telegram chats.

Code from anywhere.

Unlimited free AI image generator that runs locally by proxying public image generation services through browser automation—no APIs, no subscriptions, no credits, and no local GPU required.

The Fable Method: a structured problem-solving loop that gives any model Fable 5 grade thinking. Four Hermes Agent skills (think/act/prove/grow) + adversarial eval.

B2C e-commerce storefront powered by ERPNext and Frappe Builder (work in progress, not production ready)

PVE 集群扫描与管理平台 — Django 5 + Vue 3 全栈，Agent 采集 + 多集群管理 + AI 运维助手

Switch between AI coding agents without losing your work or context.

Find and fix what the MCP 2026-07-28 spec revision breaks in your server. 21 rules, 5 autofixers, and a readiness board.

AI Aerial Surveillance & Response System — a software-only surveillance brain that turns a closed WiFi FPV toy drone into an autonomous aerial sensor: video ingest, image enhancement, on-device YOLO detection, event logging, live dashboard, and zone-breach alerting.

OpenAI-compatible gateway for NVIDIA Triton and vLLM with tools, multimodal inputs, embeddings, reranking, and observability.

KALKI AI (Krishna Artificial Lattice Keystone Intelligence) is an Enterprise Intelligence Operating System (IOS) combining LLMs, VLMs, autonomous multi-agents, hybrid RAG, and defensive cybersecurity.

Agent Skills - 44 skills for AI agents (Sisyphus, opencode): frontend-perfection (Lighthouse audit), secret-scanner, security-review, version-bumper, commit-lint, coverage-analyzer, api-contract-testing, seo-toolkit, sql-helper, docs-system, code-review, github-repo-hygiene. skill.json manifests, CI, Conventional Commits.

Local-first memory for AI-assisted coding. Git shows what changed. Backstory shows why.

Skill router and context picker for Claude Code, Cursor, and Aider. Auto-picks the right skill for every prompt, with prompt-injection filtering for MCP.

Production-ready templates for Cursor AI — DevOps, Backend, Automation. One CLI command, full stack in 20 seconds.

本地优先的视频字幕 OCR 工具：提取硬字幕/软字幕，自动定位字幕区域，导出 SRT/TXT，并可选用大模型修正文案。

Local-first, self-hostable security-questionnaire automation: grounded, cited answers from your own knowledge base. Bring your own model — your data never leaves the host.

Mine reusable skills from your coding agents' transcripts (Claude Code, Codex CLI, or any agent) with outcome-weighted scoring and a local review queue

🔨 The meta-cognitive framework that turns any AI agent into a disciplined engineer. Iterative Reverse-engineering & Optimization Nexus.

Autonomous agent execution infinite loop and hallucination detector

ArcVellum：面向长篇小说、剧本与伪记录作品的本地 Agent 创作平台，提供文学工程硬门禁、正式正文阅读器、主动创作顾问与叙事观测仪。

Self-hosted "guess the intro" music quiz for family game night — plays your own Navidrome library on a cast TV or speaker, phones are the buzzers. Difficulty from your play counts + Last.fm. No cloud, no ads, MIT.

Opt-in Codex Skill for practical coding and externally anchored delivery with elastic agent teams, task DAGs, isolated candidates, and one canonical writer.

Control plane for AI coding agents: route tasks, reduce token spend, run multi-agent workflows, fallback executors, and track cost per task.

Baton coordinates coding agents. Each worker handles a focused task, and one orchestrator checks the result.

AI-powered university mentoring: a deterministic 8-signal risk engine ranks every mentee, an LLM narrates — briefings, summaries, meeting prep, grounded chat. FastAPI + Next.js 14.

Agent says done. VALP asks for proof.

replace your C compiler with Claude

Scores whether an autonomous AI agent actually did the work or just hallucinated its report, by checking every claim against recorded API calls. The core engine behind NoHalu.

Production-grade Agentic Harness for autonomous AI agents. Features MCP server integration, cost-controlled DAG workflows, and actor-critic loops. Optimized for Claude Code, GPT-4o, and local LLMs. The infrastructure for reliable Agentic AI.

Autonomous AI security engineer — multi-agent swarm that scans, exploits, fixes, verifies, and attests. Open source (Apache 2.0).

GD · DentOS — 开源口腔/牙科诊所管理系统:患者档案/预约/电子病历/收费/库存/消毒追溯,单机+局域网,不上云,数据主权在诊所自己手里

Push-to-talk PWA family intercom for Home Assistant — works with any smart speaker.

Self-hosted multi-platform Git activity dashboard (GitLab/GitHub/Gitee) with built-in AI code review. Aggregates pushes, commits, diffs and code stats into one page. Flask + SQLite, single-file.

AI 驱动的自动化 SRC / 渗透测试辅助平台：FOFA 资产搜集 + LLM 多 Worker 自主挖掘、审核与情报沉淀。单进程部署，仅供授权测试与研究。

AI-powered GitHub repository maintenance agent that analyzes repositories, generates meaningful improvements with LLMs, and automatically creates commits and pull requests.

A building is a program — typed, verifiable IR for Autodesk Revit

Security research on Liferay CE 7.0.3 GA4: pre-auth RCE as root (CVE-2020-7961 class) reproduced end-to-end, plus 16 more findings — 8+ with no known CVE. Agentic loop-hunt: 25 generators, 22 judges, 9 live validators on Docker. Evidence trail + one-go checker included.

Local-first terminal mission control for Claude Code, OpenAI Codex CLI, Gemini CLI and OpenCode. Run AI coding agents in parallel tmux sessions, manage direct SSH connections, monitor usage and continue securely from phone via Tailscale. No ComandOS cloud, hosted database, telemetry or SSH proxy.