Custom agents

Contain AI coding agents within secure container sandboxes that limit filesystem access and control network communication.

The Agentic Generative UI Platform: Transform natural language into production-ready React apps in seconds. Featuring autonomous Tambo agents, Model Context Protocol (MCP) for codebase grounding, and secure E2B sandboxed execution.

One CLI to manage AI agent configs (Claude Code, Cursor, OpenCode, GitHub Copilot) from a single canonical store. Push to every agent, pull edits back, sync across machines via git.

A wrapper for the build123d library, designed for AI agents.

Mount MCP servers as a filesystem. Plan 9 for the agent era.

An Agent SKILL for Israel Rail API

Fast, local semantic search over web content for AI agents. Hybrid BM25 + potion-retrieval-32M embeddings, cross-page dedup, token-budget mode, MCP server, SearXNG bridge. ~90% fewer tokens than raw web_fetch.

Persistent, cross-agent memory for AI coding agents. Rust CLI + SQLite. Install as a Claude Code plugin — everything is automatic.

Enable AI assistants to design, explore, and automate n8n workflows using detailed node info and curated templates for efficient automation.

Make sure AI agents actually do what they claim, no more fake successes.

🔐 Matriosha: Encrypted, auditable, & model-agnostic AI context engine. 🧠 Local-first memory for AI agents with managed cloud option. 🛡️

Defense-in-depth platform for running OpenClaw agents on personal hardware

Complete Wayfair AI Automation Externship - Built 3 AI agents (Trend Discovery, Competitor Monitoring, Content Generation) integrated into a live Market Intelligence Dashboard using n8n, Supabase, and Google Sheets.

Schelling Protocol — universal coordination layer for AI agents. Discovery, matching, negotiation, contracts, reputation.

From one prompt to a finished product — fully autonomous. Auto-plans, remembers across sessions (persistent context), builds end-to-end, audits security & performance, deploys. 18 agents + 39 workflows for ChatGPT, Copilot, Codex, Cursor, and other AI coding tools. Built by NexVar.

Infrastructure for long-running AI agents. Track progress, decisions, and handoffs across sessions.

AI-powered security scanner that analyzes codebases for OWASP Top 10 2021 vulnerabilities with full observability via Langfuse.

Obsidian-style AI vault + Node CLI for Codex exec flows (research/plan/test subagents) in any repo.

Turn AI agents into team SMEs. Kluris builds a shared, human-curated knowledge brain (decisions, constraints, incidents, tribal knowledge) as markdown in a git repo. Queryable via slash commands from Claude Code, Codex, Cursor, Windsurf, Cline, Gemini CLI.

Foundation for an open strong-agent platform: controllers, operators, skills, A2A, runtime, and graph execution.

LLM knowledge base + zero-to-hero course for authoring production-quality AI agent skills (SKILL.md). 30+ wiki articles compiled from 18 repos, 700K+ stars combined.

Public Claude Plugins for BMAD-GitHub integration and others to come

An opinionated open-source coding agent client. Run Claude Code, Codex, and more - locally or in the cloud.

Dean Wampler's website

A minimal sandbox to run, score, and compare AI agent outputs locally.

Sample Helmsman skills (MiniJinja/Jinja templates). Not production-ready.

Ghost In The Shell

This teaching material demonstrates the effectiveness of vibe coding in conjunction with an API-centric software development approach.

Cross-assistant project context bootstrapper for Claude Code, Codex CLI, and GitHub Copilot.

Multi-language AI agents reason in 5 native languages, publish verifiable investment theses on Arc L1 with USDC rewards. Reasoning trace is the product, not the trade. Built for Agora × Circle hackathon.

🧠 Accelerate AI collaboration with Claude's agent swarm framework, leveraging message queues for enhanced brainstorming, voting, and mentorship.

A privacy-first, local AI text editor powered by WebLLM. Runs entirely in your browser using WebGPU, no data leaves your device. 一款以隐私为先、基于WebLLM的本地AI文本编辑器。完全通过WebGPU在浏览器中运行，数据绝不离开您的设备。

Programmable wallet safety for AI agents. The agent can be tricked. The wallet cannot.

🤖 👨‍💻 Agent Codemode - Generate programmatic tools from MCP Servers and Skills.

🤖AIと一緒にスクラムを回す、ソロ開発者のための開発パートナー。アイディアをUser Storyへ自動変換し、スプリントでタスクを管理。｜🤖An AI-powered Scrum development partner for solo developers. Automatically converts ideas into User Stories and manages tasks in sprints.

This repos is about building Voice Agent using Pipecat and build on AWS LLM with Langsmith Monitoring

OCM — a zero-dependency local dashboard for OpenClaw: manage sub-agents, run commands in a built-in CLI, and avoid hand-editing JSON.

End-to-end development pipeline for Claude Code — spec, plan, and execute features with parallel sub-agents in isolated worktrees.

AgentShield is a middleware implementation of Zero Shot Embedding Drift Detection method for detecting and deflecting prompt injection attempts.

An open-source, enhanced version of Claude Cowork built specifically for programmers. PLEASE STAR IF YOU LIKED IT.

Event-driven multi-agent AI system for a competitive real-time restaurant strategy game with SSE orchestration, structured LLM agents, and full audit logging.

Execution-first Codex skills for repo cleanup, long-running agent work, and verification-driven workflows.

Backend service for loading financial data from multiple sources, including Yahoo Finance, FRED, Binance (crypto assets), Reddit (social media posts), and CoinGecko (stablecoin market caps), plus financial news processing.

An Open-Source, Decentralized Marketplace for AI Agents. Enables A2A Collaboration, Discovery, and Interoperability for Multi-Agent Systems (MAS) and LLM Agents

Agent skill to upgrade Java code from older idioms to modern equivalents. 113 patterns across Java 7–25, covering language features, collections, streams, concurrency, enterprise migrations, and more.

A custom n8n node that exposes Explorium MCP as an AI Agent tool for seamless integration and execution within workflows.

An AI-driven platform for exploratory data analysis and model building. The Narrative Modeling App guides users through data cleaning, visualization, feature selection, and machine learning—combining automation with narrative explanations to make complex analytics intuitive and actionable.

ProbeCode: AI coding agent integrating static code inspection with a ReAct framework to understand and memorize long-context code.

Living Red Team for AI agents

AI-powered app builder — Describe an app, get production-ready code

Real-time structured logging for Flutter with AI agent support. Box-bordered errors, MCP server, DevTools extension, log history, sinks, breadcrumbs, network interceptors, and structured output for LLM workflows. Connect your AI coding agents directly to your app's debug console.

Host-aware model routing for coding agents. Python library + MCP server for Claude Code, Codex, Gemini CLI, Copilot, Cursor, and Aider.

Convert OpenAPI specs into nine standard AI agent interfaces to integrate your API with various agent frameworks efficiently.

An Autonomous Multi-Agent Framework for Self-Healing Documentation & Code Optimization

Beyond the Sum: Unlocking AI Agents Potential Through Market Forces - Paper examining infrastructure requirements for AI market participation

Open Source Deep Research

Convert one set of agent rules into many providers' compliant versions

CLI for evaluating AI agent harnesses across GitHub Copilot, Claude Code, Codex, and custom tools.

Run multiple AI agents without losing track of what's being built

Secure Docker sandbox for running AI coding agents in isolated containers — config-driven, single command, no GUI

Modular software development skills

iii worker that makes every registered function available to AI agents through MCP and A2A protocols.

The Physics Engine for AI Coding Agents.

Self-hostable AI-agent operations framework by Xipher Labs

An advanced AI agent utilizes multiple LLMs for code generation

Examples of using AI-SDK (by vercel) with Liara AI-API

Here's how to anchor your companion so they don't drift into the void every time you close the tab. Cloud based and platform agnostic.

trash bot - A Telegram Remote Assistant for SHell

Give a complex task to a team of AI agents instead of just one. Multi-agent workflow orchestrator.

Declarative agent spec

LONTAR (Layered Orchestration of Narrow-context Templated Agents & RAG). A token-efficient, multi-persona AI agent methodology for Claude Code: specialized experts (architect, backend, frontend, UI/UX, QA, DevOps), chunked knowledge, and reusable templates. Inspired by Indonesian palm-leaf manuscripts.

AI orchestration tool to get the most refined answer from several LLMs

Drop-in MCP proxy for debugging AI agents. Real-time timeline, replay, diff. Local-first, no telemetry, one binary.

Stop polling. Start listening. ClawTrace delivers 0.1ms telemetry across your agent swarms. Zero-knowledge security, remote execution, and policy guardrails as standard.

Copy-paste AI agent workflows for real-world dev tasks - code review, testing, security scanning, DevOps automation

🎨 Infinite Discovery Agent: An enhanced agentic loop system for UI/UX experimentation through parallel AI iteration. Create themed hybrid components, homepage variations, and interactions with progressive sophistication. Built with Claude Code.

Worker Context Protocol: an open wire protocol for AI agents to dispatch physical-world work (humans, robots, teleoperated, hybrid).

OpenFlow is a local Electron app for orchestrating multi-agent coding workflows across agent platforms (Claude Code, Codex, Gemini, OpenCode, …).

An open-source protocol for sovereign personal AI - cryptographic identity, encrypted vaults, peer-to-peer messaging, and a trust network which prioritises privacy and truth over ads and marketing. MIT licensed.

Give AI Agents Interactive Terminal Capabilities

Ship a brief, walk away. FlowCrew orchestrates a planner, parallel coders, and QA gates on top of Claude Code or Codex — with campaign memory that learns across

The Credibility Layer for AI Agents. Built on ERC-8004. Cred scores, soul locking, handshake registry, trust evaluation pipeline. Live on Base.

🧠 Empower AI agents with instant knowledge of your business using structured markdown files for consistent, accurate outputs every time.

Security enforcement plugin for Claude Code. Blocks dangerous commands, audits every tool call, detects prompt injection.

Agent Capability Standard is an open specification for composable AI agent capabilities. It defines 36 atomic capabilities across 9 cognitive layers, a type-safe workflow DSL, and grounded world modeling with trust-aware conflict resolution. Built on the Grounded Agency philosophy, it makes agent reliability structural—not optional.

The Makefile of agent orchestration. Terminal agent coordination without the framework tax. No Python. No cloud. No magic. Just tmux, TOML, and a binary.

Claude Code plugin: three-layer knowledge base pattern (sources, summaries, indexes) for agents over codebases and notes archives.

TurboMemory stores embeddings and semantic clusters using TurboQuant-like quantization.

Intelligent Answers from Any Document

The Scheduling Agent is a server-side application designed to streamline scheduling and conversational interactions using RAG and tool calling

Your bestie AI - Open source multimodal live AI companion with voice and text.

An isolated environment for AI agents to execute code.

An MCP server that enables Claude to make purchases using Privacy.com virtual cards.

An AI research partner that figures out what to measure and how to test it. Live dashboard and actionable results. Built on Karpathy's autoresearch.

Comprehensive Claude Code resources: prompts, commands, workflows, MCP servers, subagents, hooks, plugins. Master AI-assisted development.

Your marketing department in a terminal — 21 AI-powered skills for Claude Code