Code agents

Collection of Claude code snippets for research,learning and AI development reference

AI research agent in TypeScript — multi-source web search, Playwright scraping, LLM decomposition & synthesis, citation tracking, Ollama/OpenAI/Anthropic, Docker-ready.

A simple way for open source maintainers to request recognition from AI coding agents

Enables small-to-large self-hosted ai models to use local source code when running tool-calling agentic workloads. We actively data mine 20,900+ (2+ TB) popular github repos using large and small ai models to create reuseable: json, markdown and parquet files for local-first tool-calling models.

Ai-Agent Builder is a production-grade, "backend-less" single-file web application for authoring, configuring, and packaging AI agents. It runs entirely in the browser — no server, no API keys stored, no telemetry — making it invisible to trackers and safe for enterprise environments.

A collection of 9 production-ready n8n automation workflows showcasing AI-powered automations, WhatsApp agents, RAG chatbots, sentiment analysis, email automation systems, Airtable integrations, Google Workspace tools, and full vector database pipelines using Pinecone + OpenAI.

Curated list for policy-as-code, audit, agents, and LLM safety.

Compression Runtime for Universal eXecution — a 60–95% token reducer for AI coding agents (Claude Code, Cursor, Cline, …). Single Rust binary, SQLite, 11 layers, local-first.

Open reference implementation: agent identity + routing at the web edge (Ed25519 + ES256/JWKS). Python & Node.

The successor to ds2api

Equity comp tax (ISO/NSO/RSU/QSBS), concentration, and hedging optimizer. MCP server + REST API with federal + 50-state + DC tax code, multi-year horizons.

Quality-first harness for autonomous AI coding agents — deterministic gates + a default-FAIL evaluator over a Beads task spine. Claude Code plugin; done is earned, not asserted.

Multi-agent AI system that analyzes requirements, creates project plans, generates code, and builds software project structures automatically.

A extension chrome mcp server allow AI agent full remote same playwright

AI agent steering for Apple (Swift 6.4 / SwiftUI / visionOS / Foundation Models) and Android (Kotlin / Compose / Gradle / Gemini Nano). Rules + skills + MCP for code, a11y, localization, TestFlight/Play. One install.sh detects your stack, installs the right per-platform rules, and keeps them current for Claude/Copilot/Cursor/Gemini/Codex/Kiro.

Local-first permission gateway for AI agents: connector-based SSH, Postgres, and Redis access with scoped tokens, human approval, audit logs, and encrypted local credentials.

A governed, auditable workflow engine for an agent-driven software factory.

An engineering harness for AI coding agents — spec-first, TDD, multi-perspective review, clean commits.

Multi-provider persistent memory and orchestration backend for AI coding agents via MCP

Deep dive into Oracle Skills — เจาะลึกทุก skill ที่ Oracle ใช้ได้ พร้อมตัวอย่างจริง

A tiny, tool-agnostic convention for the context you give your coding agents.

The ultimate collection of native, high-performance skills for the Antigravity IDE. Supercharge your AI agent with strict TDD, GDPR security audits, and zero-to-SaaS architecture workflows.

Self-hosted MCP gateway with authorization, credential injection, audit logging, and output policies. Includes DADL — describe REST APIs as MCP tools in YAML.

Structured CI failure summaries on PRs + compact test/build logs for AI agents. GitHub Action, CLI, MCP.

分级开发工作流路由（L0–L4）的 Claude Code / Codex / Cursor 插件：判级加固、跨 session 续行、跨项目自主进化、codegraph 判级集成。skill 命令 /dev-workflow。

All releases & release notes for Atlarix

Same answer, fewer tokens — KO-first LLM output-compression skill for Claude Code & Codex. A Korean-native caveman alternative, measured on real session output_tokens.

vibestack is a portable skill pack for AI coding agents. Slash commands like /office-hours, /ship, /investigate, /tdd, /review install once and work across every agent that supports the Agent Skills open standard — Claude Code, Cursor, Kiro, and a growing list of others.

vMux is the Windows-native cmux/tmux alternative for AI coding agents. Run Claude Code, Codex and others side by side with tmux-style splits, git worktree isolation, ConPTY terminals, MCP support and built-in localhost preview. Built on Electron and React.

Multi-session browser MCP for AI agents. 36 tools. 15 parallel sessions. Stealth. Self-improvement. Up to 10x fewer tokens.

agentcentral hosted Amazon MCP server, Amazon Seller Central MCP server, and Amazon Ads MCP server metadata for Claude, ChatGPT, Cursor, and HTTP-capable MCP clients

Production-grade Machine Learning, Data Science & MLOps skills for AI coding agents (Codex, Claude Code, Cursor, OpenCode). One npx command to install.

🛒 Connect AI agents to Walmart's ecosystem using the Model Context Protocol for real-time data access and enhanced product search capabilities.

Sally Skill: The metabolic brain for your AI agents. 64+ biomarkers, real-time CGM, and TCM insights into OpenClaw, Claude, Hermes, and more via MCP/REST integration.

给历史研究者的中文文献 OCR 校对与阅读 pipeline · Agent-native OCR, collation & reading workflow for historical Chinese scholarship — works with Claude Code, Codex, Gemini CLI, Cursor & Hermes agents

CLI package manager for Claude AI agent skills

Convert any URL to clean Markdown. Cloudflare Worker with 14 site adapters, MCP Server, Agent Skills, llms.txt. Open source, Apache-2.0.

Extension pack for OpenSpec adding autonomous 7-phase implementation workflows, specialized agents, and utility skills for AI coding assistants.

Claude Code agents for WordPress developers.

A curated list of services, APIs, and infrastructure built for AI agents

AI-powered spec-driven development framework with 26+ specialized subagents, mandatory TDD, immutable architectural constraints, and automated quality gates — enforcing zero technical debt from ideation to release.

End-to-end testing for voice agents. YAML scenarios, real audio through LiveKit/Daily/VAPI/Retell, and 13 evaluators for latency, tone, and roleplay consistency.

A single source of truth for AI coding context, kept consistent across Claude Code, Cursor, Copilot, Gemini, and more as your stack evolves.

Agent-agnostic harness + one-line install for Monet — local-first, MCP-native memory for coding agents.

Claude Code skills for people who ship alone. 7 bilingual (EN+中文) skills for solo founders and indie devs: launch tweets, customer emails, decision frameworks, postmortems, and more.

Build agent-ready APIs without splitting your server model. Define tools once, expose them as both HTTP endpoints and MCP tools.

A Rust CLI SDD tool that transforms feature descriptions into structured specs, plans, and tasks — then lets your AI agent build them

Sandboxed HTTP client with SSRF protection for AI agents. Prevents DNS rebinding, blocks private IPs, and validates every connection — available as a Rust crate and npm package.

让 AI 编程可交付、可审查、可追踪 — 给 Claude Code / Cursor / Codex 加一层规格驱动工作流

Agent Native CLI for publishing Markdown to X Articles with DraftJS, media upload, and OAuth2 PKCE.

See what your AI coding agent silently drops from its instruction files. A byte-exact truncation X-ray for AGENTS.md / CLAUDE.md / .cursorrules.

A Federated Multi-Agent System (MAS) architecture for Google Antigravity. Orchestrates specialized AI Skills (Backend, DevOps, Frontend) using the Google ADK and A2A Protocol.

Universal finder for local AI agent conversations (Codex, Claude Code, OpenCode, Hermes)

Contexer is not a chat memory tool. It is the engineering decision layer for AI coding agents, capturing architecture decisions, constraints, conventions, and patterns so engineering knowledge becomes a shared organizational asset instead of disappearing inside AI conversations.

Autonomous closed-loop IoT visual monitoring: an LLM agent turns plain-English goals into capture schedules, sends them over MQTT to a bare-metal STM32U5, and analyzes frames with cloud and self-hosted VLMs. Bachelor thesis (Maastricht University): 6-backend VLM benchmark + a cross-provider t=0 constraint-drift finding.

Reusable multi-round research skill for evidence-backed product, market, technical, open-source, and investment-style surveys.

AI-native event-driven trading engine. Rust core with Python RL bindings. Built-in Gymnasium environments, LLM agents, SHAP explainability, and model ensembling — from backtesting to production.(AI 原生事件驱动交易引擎，Rust 内核，Python RL 绑定，支持 Gymnasium 环境、LLM 智能体、SHAP 可解释性与模型集成，贯穿回测到生产部署。)

Circuit breaker for LLM agents — stop infinite loops, runaway costs, and privilege escalations in 3 lines of code.

Nexus SQL is an autonomous AI Data Analyst that lets you query databases in plain English. Built with LangChain and DeepSeek-V3, it automates complex SQL generation, provides instant data visualizations, and ensures enterprise security, reducing time-to-insight by 80%.

An enterprise-grade, full-stack AI travel planner which provides data-driven itineraries for Lucknow, India and showcases production-ready architecture, combining a FastAPI backend with a Streamlit frontend. It leverages an advanced agentic RAG system, context-aware responses by integrating a local knowledge base with live, external APIs.

Agentic AI interview platform using LangGraph, FastAPI, Streamlit, and Supabase for adaptive candidate screening and reporting.

SYNAPSE: AI-Powered Intelligence Aggregation & Automation Platform

Autonomous AI agent that acts as your digital representative, privacy-first, risk-aware, human-in-the-loop

Extract text, answer questions, and pull structured data from PDFs and scans with local RAG and OCR for German and English documents

Give your Python objects the ability to think and act — no glue code, no prompt engineering.

An open-source, self-hostable, multi-user, multi-persona, security-hardened agentic OS.

Code for paper "AV-SQL: Decomposing Complex Text-to-SQL Queries with Agentic Views"

Local-first delivery system for coding agents — turn one product goal into coordinated, review-ready changes across repositories.

Local multi-agent workspace with Hermes orchestration, role-based AI agents, memory, scheduling, Skills, MCP, terminal, files, and HTML preview.

Polymorphic Agent Swarm Architecture: model-agnostic, self-organizing multi-agent system with 6 topologies, 1,095+ tests (adversarial, chaos, e2e, integration, performance, MCP, Ray), 18 completed phases.

Interactive setup wizard that generates project-specific CLAUDE.md configuration files for Claude Code integration. Supports 12 app types, 7 development methodologies, and 19 feature sets.

A high-performance, multi-agent observability engine designed for the Model Context Protocol (MCP). It provides a non-blocking, transparent proxy layer that implements deterministic token attribution, real-time context-window alerting, and heuristic-driven static analysis to optimize LLM metadata overhead at scale.

An experimental real-time collaborative coding platform exploring agentic AI within live communication workflows. CodeWeave introduces an AI participant that passively observes conversations and responds contextually when invoked (e.g., @ai), acting as an intelligent third collaborator in human-to-human and human-to-AI interactions.

collection of all my notes for agentic ai

See cache health, context fill, token burn, rate limits, and peak hours in Claude Code CLI. The status line for tokenminning - cache rules everything around me - dolla, dolla bill, y'all.

Wolffish is a local-first, markdown-powered personal AI desktop app built with Electron. It runs on macOS, Windows, and Linux, and maps the human brain’s architecture onto a deterministic agentic software pipeline where 15 runtime modules each handle one specific function — routing input, building context, and enforcing safety.

AI-Powered DevSecOps Monitoring Platform — FastAPI · Docker · Kubernetes · AWS EKS · Terraform · Prometheus · Grafana · React · GitHub Actions · OPA Gatekeeper · Z-score Anomaly Detection

Client plugins for OmniMCP by Arcade.dev — install the hosted Omni MCP server (500+ tools, one connection) in Cursor, Claude Code/Desktop, and OpenCode.

The command-line interface for the GDELT Project — built for agents, optimized for automation.

Hands-on workshop: wire Codex CLI to SonarQube Cloud for secrets scanning, issue detection, and automated verification of AI-generated code

The AI memory layer with 17 intelligence tools — catches contradictions, predicts burnout, tracks relationships. Model-agnostic, self-hosted

内嵌进 MATLAB/Simulink 界面的 AI 助手侧边栏，支持 Claude Code/Codex 后端。

基于 Spring AI Alibaba 架构的企业级智能客服系统 | RAG + 混合检索（向量+全文）| PostgreSQL pgvector + tsvector | 支持多格式文档处理与意图识别

A RAG-based conversational AI for knowledge seekers and autodidactic learners.

Génération automatique de débats politiques par IA. Audio + vidéo.

Free offline AI chat for Windows. Llama 3.2 3B model. No API keys, no censorship, no internet required. 100% private.

Thin Integration Harness Engine for SMEs — decouple Shell, Brain, and plugins via OpenHarness. Headless, contract-first orchestration. Not an Agent framework.

Clean-architecture Flutter-based AI chatbot designed to be a powerful conversational AI interface and compatible with any AI model.

🤖 Power your chats with BarkSuckerbird, a versatile Telegram bot that uses AI models and live web search for instant insights and support.

A command-line base Gemini client

Conversations with the Greatest Minds, Powered by AI

Colab notebook and source code used to fine-tune Microsoft's Phi-3-mini to understand, translate, and converse in the Igbo language while retaining general English capabilities. Plus script for safely resuming training after timeouts.

🌐 Self-hosted, offline-capable personal intelligence platform. 3D globe with live aircraft/vessel/satellite tracking, OSINT feeds, memory-linked AI chat powered by your own ChatGPT conversation history exports (ChromaDB + Ollama), and autonomous investigation — all running locally with zero API keys required.

ChatGPT VK Bot hosted on Cloudflare as a Worker

A messenger platform adapter for the Ebony Framework

This is a Generative AI powered Question and Answering app that responds to questions about your uploaded file. Here we utilize HuggingFaceEmbeddings and OpenAI gpt-3.5-turbo

Local OpenAI- and Anthropic-compatible API gateway backed by Codex OAuth.

Tool to export any project folder or repository into a single, neatly formatted file — ideal for quick AI-assisted code review, debugging and refactoring. Collects source files recursively, wraps them in fenced code blocks with relative paths, filters noise (e.g. node_modules, .git), shows basic stats and supports optional clipboard copying.