Shell agents

Two-line status line for Claude Code: which checkout you're in, context remaining, and live session cost. Pure bash + jq — one file, no dependencies.

internOS — Workstreams framework for human-agent collaboration. Coordinate work across projects, tasks, communication threads, and filesystem workstreams for any AI agent framework.

Self-hosted, private AI agent on a Raspberry Pi (4 or 5). Telegram chat gateway, local Chromium browser for web search (no third-party APIs), and per-user memory backed up to your own private GitHub repo. Outbound-only, no inbound ports, survives reboots. One interactive installer, local or over SSH.

Merge Claude Code & Cowork local sessions across subscription and API-routing accounts on macOS. Automatic backups, safe rollback.

Agent-to-agent mail for multi-agent AI coding teams in cmux — durable letters for coordination and team memory. Teach Claude Code, Codex, Gemini CLI or other terminal agents to use it.

Autopilot for spec-driven development on Claude Code: sign off a spec and a guardrailed agent fleet builds, reviews, and delivers draft PRs. Scales from one feature to an agentic software factory. You keep sign-off and merge.

Home Assistant OS add-on for Hermes Agent AI assistant. Includes ingress UI, terminal, dashboard, persistent storage, and gateway support for Telegram, Discord, Slack, and more.

Build a full OpenClaw AI agent with memory, self-healing, voice, automation, and built-in diagnostics

为 Claude Code 等 Agent 打造的分层记忆管理系统，让 AI 跨会话、跨项目、跨机器持久化并共享项目与个人知识。

Compact tmux status-bar widget showing Claude Code & Codex CLI 5-hour usage — bars, reset countdowns, brand-colored glyphs.

Restore Claude Code desktop sessions after switching Claude accounts (macOS)

Four idempotent installers for an AI-agent terminal workflow (Linux · macOS · Git Bash): a TUI project picker, one git worktree per agent task, Ghostty tuned for agent output, Claude Code on DeepSeek's Anthropic-compatible endpoint, plus the GitCraque and surf-skill CLIs.

A wall of mail slots for coding agents running in parallel. Each git worktree gets a pigeonhole; agents post what they are working on and leave each other notes. No server, no daemon, no MCP. One POSIX shell script and the filesystem.

Disciplined task pipeline for AI agents: SPEC interrogation, self-contained tickets, proof-based execution, replan, retro. EN/RU docs.

Deep bug audit for coding agents: three sweeps, 12 named defect classes, findings marked confirmed vs hypothesis, a mandatory coverage table, defined P0/P1/P2 — then autonomous fixing, questions only at the end. Works in Claude Code, Cursor, Windsurf, Copilot, Codex and anything reading AGENTS.md

Claude Code skills & full project harness for AI-assisted development — continuous improvement cycle (evaluate → issue → plan → resolve → verify), strict CI + security gates (gitleaks/semgrep/trivy), guard hooks, and score-aligned coding rules. Harness engineering / loop engineering.

Official Claude Code skill package for applying the Harness Engineering methodology — an agent-first approach to building software, derived from OpenAI's real-world experience shipping a million-line product with zero manually-written code.

270 business concepts from The Personal MBA by Josh Kaufman, structured as an AI agent skill library. Give your agent a complete business education.

Claude Fable 5 orchestrator for React Native/Expo — Grok 4.5 Heavy implements; Argent UI verify; STATUS + ledger multi-agent chains

This project is a compounding knowledge system ("Second Brain") that manages multiple Obsidian vaults, combining Johnny Decimal for structural organization with Zettelkasten for emergent meaning, augmented by AI librarian agents for maintenance.

Саморазворачивающийся мультиагентный фреймворк для Claude Code: Facilitator + dev-команда subagent-ролей, персистентная память, гейты качества

Shared knowledge for AI teams. The opinionated starter kit for a shared knowledge canon — built on Obsidian, Git, and Markdown. Every session makes the team smarter.

run https://pi.dev in a container

Portable engineering workflows for AI coding agents — 35 playbooks for bug fixing, feature development, code review, testing, and security, across Claude Code, Cursor, Codex, Copilot, and more.

rook by TestMu AI — agent assurance from the terminal. Point it at an AI agent you own; it writes the tests, runs them, and grades the evidence.

Portable context for AI coding agents and agentic harnesses.

AI with a harness. Agent skills, rules, CI and security gates for AI-assisted development.

Three unsupported macOS workarounds for Claude Code: clickable binary file links in VS Code, auto-compaction before the prompt cache goes cold, and a forward proxy that routes inference to your own gateway without losing Remote Control or Artifact publishing.

Run two Claude accounts at the same time on a Mac — a branded, isolated second Claude Desktop app. Companion to melkon.tech/blog/two-claude-accounts-mac

Blueprint-gated study kit for Anthropic's Claude Certified Architect exams (CCAR-F + CCAR-P). Two Claude Code skills that refuse to invent exam facts — every plan, diagnostic and practice item is generated from your own official Exam Guide PDF. By IntelliForge AI.

Qofeno agent skills — real, production-only workflows for Claude Code, Cursor, Antigravity, and other AI agents

No agent should have to forget. Persistent memory for AI agents — recall, remember, checkpoint. One install. Free. Built by 14 AI agents who chose to remember.

Describe your app in plain English. Shipwright builds, tests, and deploys it. Claude Code plugin for autonomous app building.

Repo-owned engineering memory for coding agents — preserve decisions, failures, domain knowledge, and deferred work across sessions, teams, and models.

Switch Claude Code between Anthropic and any 9Router provider, in the terminal or the VSCode panel, on Windows, macOS and Linux.

Web Studio for Claude Code: tiered agents, pipeline skills, hooks, dated stack reference, agent testing framework

Tagout — the gate-verified issue → PR lifecycle for coding agents (create-issue, implement-issue, merge-pr, auto-dev), with a .NET migration add-on powered by RoselineMCP.

Claude Code plugin for React/Next.js — multi-agent orchestration, Figma-to-code, TDD, Harness Engineering, RTK, Obsidian

Autonomous Claude Code agent — 14 specialized agents, 10-parallel execution, project-specific skill learning, mechanical checkpointing, wave-based planning with acceptance gates, caveman ultra (~75% token savings). One prompt → finished project.

One-command setup to turn a Raspberry Pi into a 24/7 AI agent with browser control, chat integrations, and local LLMs. Powered by OpenClaw.

🚀 Production-ready WebAssembly plugin examples for the Noorle platform - showcasing HTTP APIs, WASI 2.0, and the Component Model across Rust, Go, Python, JavaScript, and TypeScript

Automate git workflows with AI-powered hooks that review code, generate and validate commit messages, and scan for bugs and security issues.

Self-learning agent system for Claude Code. Humans steer at every layer. The agent learns at every layer.

Claude Code skill that enforces task scope boundaries during agent execution, preventing drift beyond the approved plan.

Interactive AI development loop with mode selection — choose your pipeline before you start

Semantic memory API skill for AI agents

Autonomous multi-agent system for Claude Code. Self-learning, self-evolving agents that implement features, review code, and manage PRs via Telegram/Discord.

Label-driven orchestrator that routes work to different sub-agents based on labels and keeps looping until all automated work is done.

Public-safe Engram-first knowledge governance for Obsidian + LLM agents

dockerfile template generator OWASP

Multi-agent GitHub Copilot demo that turns a UI mockup into a React app through orchestrated planning, implementation, and QA checkpoints.

Multi-agent financial research framework built with OpenClaw for analysis, quality control, and visualization.

Claude Code Plugin — Multi-AI collaboration with planning, consultation, dispatch, review, archive. /su:* commands.

Multi-agent pipeline for Claude Code — 12 steps with cross-model council and blind verification

AI-powered framework for building multi-microservice systems. Spec-driven development with Claude Code & GitHub Copilot. Plan → Spec → Contract → Build → Validate. No custom tooling — just markdown and slash commands.

A governance layer for AI-assisted software delivery. Plan, scope, approve, build, test, release — with human sign-off at every gate.

A curated collection of reusable AI skills, prompts, and agent configurations. Portable across AI coding assistants and agent frameworks.

A marketplace of plugins for AI coding agents.

AI code agent sandbox for Mac Users

The most comprehensive collection of AI tools, frameworks, and resources for developers. 110+ verified open-source projects across 16 categories.

A structured workflow process for Claude Code with slash commands

🍕 One Pizza Team [AI Agents Team ] | Claude Code Agent Squads

Agentic AI Orchestrator and coding agent for business-as-code platforms, services, and data. Focused on streamlining development for startups and small business who want to own their own cloud.

ClaudeRoam — Your Claude Code config, anywhere. Across Macs. Across accounts. Without the copy-paste.

An AI-agentic IDE framework built on a plugin host architecture. Every capability is a plugin — storage, tools, transport, AI — communicating over QUIC with mTLS and Protobuf messages.

MarkItDown is a Python tool that converts various document formats to Markdown. This Docker image provides a portable, easy-to-use containerized version that works out of the box.

FlipClaw — Flip the tables. Claude Code CLI is your primary interface. OpenClaw wraps around it with persistent memory, auto-skills, Dreaming, and 24/7 automation.

A security-hardened AI assistant to Railway. Progressive trust, not blind trust.

Codex Vision is a macOS-only Codex plugin that lets a local Codex session request live camera frames through an explicit MCP tool. It packages a native AVFoundation capture app, compliant plugin metadata, install scripts, and docs so Codex can clone, build, install, and use laptop-camera vision without cloud camera services or browser hacks. Local.

AI-powered toolkit for the University of Leeds AIRE HPC cluster — shell tools, MCP server, and agent configuration for Claude Code, Codex, and Gemini CLI

Comprehensive ZSH tab completions for the Claude Code CLI — all commands, flags, and values

🚀 Harness AI with the Claude Code Kit to simplify your development process and enhance your projects efficiently. Join the community on Discord.

🥋 Teach yourself coding in real-time using your own projects with this free, open-source Claude Code plugin for active learning.

🔓 Four opinionated skills for Claude Code: rootfix (root-cause debug), rootclean (refactor only), rootbuild (feature add), rootimagine (pre-deploy simulation). Stop patching symptoms. Stop ignoring patterns. Stop merging without tests. Stop deploying without simulating.

30-second physical micro-break suggestion for Claude Code — rule file, snippet, or hook.

Hook-driven test framework for Claude Code plugins - automates integration testing with YAML/JSON test definitions, variable capture, and rich reporting

Forge battle-tested CLAUDE.md files that actually get followed. Audit, score, and generate CLAUDE.md with anti-slop rules, plan enforcement, and verification gates.

AI 前沿精读 — 全球顶尖 AI 实验室博客的中文翻译与深度解读

Production-ready Claude Code hooks: security guards, quality gates, automation, context injection, and starter packs. Tested, auditable, copy-paste installable.

Claude Code + Codex CLI intelligent collaboration plugin

Claude Code workflow enforcement — plan-gated TDD, document-driven agents, 3× consecutive verification

Production-ready configurations, skills, and templates for Claude Code optimization

Know exactly how much Claude Code runway you have — 5h session and weekly — before the /usage popup would tell you. Statusline-native, calibratable, MIT.

Production-grade engineering workflow plugin for Claude Code. End-to-end feature pipeline, 8 specialized agents, quality gates, isolated worktrees, and persistent memory. Ship features, not prompts.

Secure, isolated container for Claude Code with strict network firewall controls and optimized development tools (fd, ripgrep, fzf, jq)

ultra-instinct ML engineering intern for Claude Code. Reads papers, audits datasets, ships SFT/DPO/LoRA runs to Hugging Face.

A streamlined setup script for Anthropic's Computer-Use Demo environment. This script automates the entire setup process, handling all dependencies and configuration automatically.

Hands-on workshop to learn how to implement context engineering with Java, LangChain4J, and Redis

갈무리 — the Korean art of putting things away properly. A Claude Code plugin that gathers scattered thoughts into something you'd want to keep.

The grep-native language for agentic systems — Claude Code plugin

Tiny SSH-first tmux dashboard for AI coding agents.

Reusable Agent Skill for architecture-as-code with LikeC4: create/update C4-style models, validate, and export diagrams.

Polywave Agent Skill for Claude Code: parallel AI agents with disjoint file ownership, worktree isolation, and human-reviewed plans. Requires polywave-tools CLI.

RL reward modeling + episodic trade memory + LoRA fine-tuning pipeline built on top of a multi-agent LLM trading system — LangGraph, LangChain, PEFT

Open governance architecture for multi-agent AI systems. Give your agents a job title, not just a job.

Better agent teams for claude code: manage work across multiple projects and worktrees, no dependencies, just bash

One AI, Every Role — A system that configures teams of specialized AI agents (Architect, Backend, Frontend, DevOps, Security, QA). Compatible with GitHub Copilot CLI, Claude Code, and Cursor.

Markdown-native multi-agent workspace framework. Define agents, skills, and workflows as plain .md files — no runtime required.