Code agents

MCP server for AI agents to record browser demos with animated cursor, auto-generated video, screenshots, and documentation

A high-performance desktop app interaction framework for AI agents — faster, cheaper, more accurate. / 让 Agent 操作桌面软件更快、更准、更省。

Agente de excelencia para OpenCode que crece contigo. 9 fases, 11 skills, auto-adaptación y aprendizaje de errores. Cuanto más lo usas, mejor te entiende.

Relationship Persona Simulation Engine — 5-layer personality model, nonlinear state dynamics, LLM dialogue analysis. Build a digital twin from chat history. 不是替代真实的人，而是帮你更理解人心。

A culinary agent for Claude Code: design and cook a dish through 7 phases, grounded in regional cooking knowledge with dynamic ingredient curation.

Multi-model panel and judge (OpenRouter-Fusion style) for Claude Code, opencode, and any MCP host. Diverse models answer in parallel, then a judge reconciles them into one better answer.

Servidor MCP para retail VTEX: busca, agrega al carrito, verifica stock por local y analiza tu gasto desde Claude Code / Cursor. Primera implementacion: Plaza Vea (Peru).

Local-first AI writing control center for long-form fiction: structured story memory, long-context prompts, consistency checks, AI-tone cleanup, and IP adaptation.

A Full-Stack Open-Source Platform for Agentic AI Research Assistance via Deeply-Engineered Multi-Paradigm RAG

破窗造视 Fractovision — MiniMax模型多媒体生成引擎。图片+视频+语音+音乐四合一API封装。

VEXI is an open-source AI coding agent for the terminal. Bring your own API keys, work with OpenAI, Anthropic, and other providers, and code faster with a lightweight, multilingual, privacy-friendly CLI.

A slim, local-LLM-first AI coding assistant. TUI, Web UI, and desktop app. Runs air-gapped with Ollama, vLLM, or any OpenAI-compatible endpoint.

Your autonomous AI coding agent. Open-source. Local-first. Fully extensible.

Dynamic Effort Orchestration — a Claude Code + Fable 5 harness that assigns each sub-agent the effort level its task deserves. Zero Opus fallback.

Enterprise-grade schema drift detection for Microsoft Fabric with automated change detection, impact analysis, schema history, and DevOps integration.

Self-hosted runtime for personal apps that combine reliable code with LLM agents and adapt across every screen.

Scan your LLM-powered applications for authorization gaps, leaked credentials, missing rate limits, prompt injection risks, and other security issues — before they reach production.

AI-powered English learning platform with FastAPI, Next.js, LangChain, OpenAI, and Azure integrations for personalised assessments, adaptive exercises, speech analysis, and intelligent feedback generation.

[STUDY] Application that turns into a Chat-Bot by entering the OpenAI API Key and Prompt.

Self-hosted multi-tenant LLM gateway: adaptive routing, multi-agent prompt-injection defense, semantic caching, and a closed-loop cost optimizer. Java 25 · Spring Boot 4 · Spring AI · LangGraph · pgvector · Redis · ONNX · Prometheus · Grafana · Next.js · Docker. Ships a benchmark + a paper.

Aiyu Multi-Agent - Lightweight AI Agent Framework with Tools, Plugins, and Multi-Agent Orchestration

智驭万象 · Cosmosys — 全球首款端到端多智能体控制自动化平台。Natural Language → Controller Code.

Persistent ground truth for Claude Code. Stop AI agents from drifting on your codebase.

AI-powered autonomous business platform with Ray, FastAPI, and Streamlit — marketing automation, content generation, and e-commerce

Watches live drilling data, flags trouble before it becomes an incident, and answers plain-English questions about the well — built entirely on Equinor's real Volve field data.

Plumbline — a self-learning, customer-value-governed agile AI agent team for Claude Code. 87 subagents + skills, TDD defense-in-depth gates, Kaizen retros, a four-body adversarial council, and an empirically benchmarked QA harness. Does it hang true?

Self-hosted multi-agent AI platform. Persistent labs, 40 sandboxed tools (code, web, RAG, GPU dispatcher, web3), private Qdrant + LightRAG, sandboxed code execution per lab, anti-loop detection, Ollama/vLLM/OpenAI/Anthropic adapters. Two-command Docker deploy. Apache 2.0.

Open-source AI workflow engine — build, run, measure, and self-improve multi-step agent workflows. macOS app, MCP protocol, CDP browser automation.

🔥 LangGraph 多智能体研报系统 | Planner-Researcher-Reviewer 评审闭环 + 自动打回迭代 | FastAPI 异步任务 + Redis | Streamlit 实时看板 | DeepSeek 驱动

Runs CLI coding agents (Claude Code, Codex, Copilot, Cursor, Gemini, opencode) against a task queue. Each works in an isolated VM; output is checked by configurable auditors, then merged via git. Pools multiple provider subscriptions with quota-aware routing and fallback. C#/.NET 10, MIT-licensed.

Run your AI coding tool through a 18-stage dev pipeline. Multi-host across Claude Code, Codex, Gemini CLI and OpenAI API compatible endpoints. Artifacts and gates on disk, not a chat log.

Pure SQLite memory system for Hermes Agent. Zero Docker, zero external services. 5-layer self-healing (v2.1+) + CIRAAF macro-restructuring. 中文 FTS5 patches included. Cross-session global sharing out of the box.

Universal AI agent skill: Fable Brain behavior + Karpathy agent-orchestration workflow + 50 production-ready prompt templates. Paste as system instructions.

Everything you need to learn Claude Code: deep notes, real assignments, working solutions. From agentic loops to parallel multi-agent systems.

This repository contains code and walk-through for creating a local context engine

A terminal control tower for all your Claude Code sessions

Voice Agent SDK helps you build realtime voice agents with prompts, tools, knowledge, providers, and guardrails, so your AI doesn’t wander into the conversation like it just got promoted by accident. Structured enough for production, flexible enough for weird ideas, and smart enough to ask before doing something expensive.

A maintainability ratchet for AI-assisted Python.

MCP design system validator for agentic coding. Detects design drift in agent-built UI.

Self-hosted AI Analytics Engineer. Bridges the gap between your codebases and databases.

Four AI engineering skills that gate your agent before it builds. Evaluated with deterministic gates, behavioral probes, and LLM-as-judge rubrics.

AVE - Agentic Vulnerability Enumeration. The behavioral classification standard for agentic AI components.

A lightweight spec-driven loop harness for Claude Code, Codex, and Gemini CLI. Fresh-context runs, specs, commits, and gates to get you and your agents productive immediately.

AI Growth Mirror 是一面本地优先的 Agentic 操作成熟度评估系统。它读取 Cursor、Claude Code、Codex 等 AI 编码工具的本机会话历史，把“协作是否稳定、问题出在哪、如何改进”转化为可观察、可解释、可追踪的成长报告与训练建议。不是 Prompt 打分器，而是帮助开发者建立可复用 AI 协作系统的成长镜。

BENCHMARKS WITHOUT BORDERS

Cantus is a Claude-first, local-first desktop coding environment for macOS (Apple Silicon). A native Tauri 2 app pairs a Monaco editor, an integrated terminal running the Claude Code CLI in a backend PTY, per-hunk and per-line git staging, resumable Claude sessions, a .claude workflow task runner, and a local SQLite-backed learned-memory store.

Claude's free extension, No Claude subscription required, just use it with openrouter api, using free models.

如何将动态变化、经常被训练或计算任务占用的本地消费级 GPU 资源，封装成一个可弹性升降档、持续可用、并对外保持 OpenAI / Anthropic API 接口稳定的大模型推理服务。

Flatten Claude Code sessions: keep every prompt and event verbatim, resume at a lower token count. An MCP server.

Self-hosted spend firewall and gateway for LLM ( OpenAI / Anthropic / Gemini ). Hard per-user & per-project budget caps that block runaway costs before the API call, plus cost-per-customer tracking, semantic caching, and failover. One line of code, single Go binary.

Lightweight token tracking, cost management, and budget enforcement for LLM API calls

Clean the AI out of text and detect AI slop - bilingual (EN+RU) Claude skill. Rewrites AI cliches into natural prose, rates AI-likeness, and tells decorative AI emoji from genuine human emoji.

Persistent code-knowledge memory for AI chats. An MCP server that caches a codebase's structure and summaries in a local SQLite "brain" so sessions query it instead of re-reading files — token-cheap, 100% local.

Audit how well Claude Code followed your CLAUDE.md — and prescribe fixes (promote to hook, delete dead rules)

A full-stack Claude Code studio — 37 agents, Studio Brain onboarding, game dev track, design phase. Bring an idea, the studio handles the rest.

Native Model Context Protocol (MCP) server for Zabbix with an embedded AI assistant. Auto-generates API tools from the installed Zabbix version, enforces RBAC, supports OpenAI/Anthropic/Gemini/Ollama.

An OpenAI-compatible LLM proxy that flocks toward the fastest provider

Keep your Claude Code prompt cache warm during breaks. Saves up to 92% on cache rebuild costs.

A simple, elegant Python library and Mini SDK for AI models with powerful features. Built for developers who want to integrate AI into their projects without dealing with complex API setup. Supports Google Gemini, OpenAI, Anthropic, and any custom provider via the Adapter Pattern.

Refine and translate selected text in any macOS app. Services menu, global hotkeys, and a floating action bar over your selection — powered by your favorite LLM provider.

修复ClaudeCode更新到最新版本后，无法使用第三方模型的问题。Fix the issue where Claude Code can't use third-party models after updating to the latest version.

Permission deny rules that stop Claude Code's built-in tools from reading your .env, secrets, and credential files.

After a Claude Code session, what did AI contribute vs what you drove? collab-proof records it before it evaporates. Zero dependencies.

Critical thinking plugin for Claude Code. Three specialized agents (advocate, adversary, judge) argue, stress-test, and judge any idea using a shared argument graph, 10 reasoning traditions, and persistent memory.

Lunar Vault is a self-contained memory infrastructure for Codex. Features zero-dependency ONNX embeddings, advanced reranking, and secure one-click installation without leaking personal data

Self-hosted multimodal RAG cloud memory service — a shared brain for your AI agents. Multi-agent cross-query, LanceDB + LightRAG + RAG-Anything.

The open-source Agent alternative — 34 tools, local RAG (FTS5+vector+RRF), Prompt Caching, MCP/LSP/Skill/Hook, all in one cross-platform desktop app. Apache-2.0.

Maxwell's personal site — maxwellii.com 终端体个人主页 + RAG-powered AI 对话化身 (Next.js 16 + 火山方舟 doubao-pro) + 简历多版本归档

Mithra Life OS is a production-grade, AI-native personal operating system that consolidates task management, habit tracking, journaling, and focus tools into one unified platform

Production-grade Deep Research AI with web search, reasoning models (OpenAI o3, DeepSeek-R1), and inference-time scaling.

AgenticNET is an open-source, Azure-native platform for building, deploying and operating enterprise AI applications with .NET. It combines intelligent routing, RAG, reviewer loops, infrastructure automation and full decision traceability into a single production-ready platform.

CI-grade evidence and safety tools for AI agents, MCP servers, and upstream contributions

160,000 deductively-derived JSON constraints that eliminate LLM persona drift. Measured: 0 breaks in a 120-turn pressure test (plain prompt: 8).

AgentCrew is a Markdown workflow for coordinating AI agent teams with roles, playbooks, skills, handoffs, and human approval gates.

AI agent for X/Twitter with full read + write capabilities. Free reads via twitter-cli, writes via official X API MCP server.

A content intake routing skill for deciding when humans should read, agents should process, or both should co-read.

Your AI agent skill doctor - 5-dimension scoring + security gate + leaderboard

Manual goal mode with persisted state, drift blocking, and proof-based completion for GitHub Copilot CLI.

Локальная мульти-агентная система на self-hosted LLM (Ollama). 3 канала (Telegram/консоль/MAX), 4 типа моделей (LLM/embed/vision/voice), мультимодальность (текст/PDF/фото/голос), роли Planner/Executor/Critic с рефлексией. Инструменты: почта, Яндекс.Диск, web, OCR, cron-планировщик. Семантическая память (sqlite-vec). Coverage 88%. Без облачных API.

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.

Pre-deploy static scanner for risky AI agent capabilities. Local-only. No telemetry. No network calls during scan. Redacted output. MIT.

Intent-addressable code for AI agents. Causari records every action an agent takes on your codebase - the prompt, model, reads, writes and reasoning - and lets you trace, diff, revert and bisect them like git commits. With bidirectional causal provenance and a built-in MCP server.

Local-first, human-in-the-loop personal AI employee. Watches email/LinkedIn, drafts actions with OpenCode, and waits for your approval via Telegram/Discord, before anything real happens.

macOS menu-bar mission-control for your Claude Code sessions — see, hear and reply to every agent so none sits idle waiting on you. Rust + Tauri, event-driven on native hooks, fully local.

Comprehensive reference notes, diagrams, codelab walkthroughs, and interview preparation material for Google's 5-Day AI Agents: Intensive Vibe Coding Course.

A zero-dependency PowerShell script to switch between multiple Anthropic Claude accounts.

Beast Agent is a TUI to an agent, much like Pi or Claude Code. But compiled. Containerized. Multi-model. Agentic. Perfect for long-horizon tasks.

Portable workflow protocol for safer multi-agent AI coding with explicit review, evidence, handoff, and git gates.

Session resources and patterns for orchestrating multi-agent workflows in VS Code — decomposing work, forking vs. delegating, and verifying quality at scale. From Microsoft Build 2026.

Production-grade skills, custom MCP servers, and real projects built with Grok Build 0.1 (grok-build-0.1). The definitive toolkit + living showcase for xAI's agentic coding model. Sole author & maintainer: Cobus Greyling

A visual workspace for designing, configuring, and managing OpenCode agent teams.

Portable Goal-Based SDLC for Claude Code — interactive goal-mode + autonomous park-and-continue loop-mode, installable into any repo as a plugin. Zero runtime deps.

Local-first RAG platform for technical document libraries. Features FastAPI, ChromaDB, and active retrieval (FLARE) powered by Ollama.

The open framework for governed AI development: EU AI Act compliance, PII redaction, MCP proxy, prompt-injection firewall. Python + Rust, Apache 2.0

Learn from any GitHub repo systematically with AI agents guiding you through architecture, flow, patterns & takeaways

Multi-agent coordination + harness adapters + portable CLI utilities for Claude Code / Cursor / Codex.

Skills + MCP server for Claude Code, Codex, Hermes, or OpenClaw - headless MIKE+ automated modelling and analysis (inspect, edit, run, read, plot), no GUI.

Let any MCP agent build, run, and debug FiveM resources live.

The modern, local-first PostgreSQL client with a natural-language SQL assistant.

Your agent wastes context. Distill fixes it.