Coding agents & developer tools
Claude Code, Codex & Cursor extensions
8,960 agents · most starred first
Claude Code, Codex & Cursor extensions, page 67 of 90
Local dashboard to switch the active Claude Code account with one click, showing each account's 5-hour and weekly quota. Zero dependencies, host and WSL, Docker-ready.
Neovim-style TUI for DeepSeek Harness (dsh): nvim renders the terminal, a Node runner bridges DSH agent events over msgpack-RPC — sessions, tool cards, reasoning panel, approvals, markdown tables
Agent skills for Cursor, Claude Code, and Codex — reliable agent loops with scope, verify, and resume
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.
⚡ macOS menu bar app showing your Claude AI usage limits in real time. Zero setup — auto-detects Claude Code credentials.
Reusable, tool-agnostic AI engineering skills, workflows, templates, and evaluation playbooks for AI coding agents.
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.
Саморазворачивающийся мультиагентный фреймворк для Claude Code: Facilitator + dev-команда subagent-ролей, персистентная память, гейты качества
75 agent skills for Hermes Agent and Paperclip: research, write, organize and operate. Every skill passes a QA gate with hard safety rails on spending, sending and deleting.
One agentic workflow, every usage pool you own: an operator Claude delegates to worker Claudes across your accounts (local MCP), with zero-context-loss failover and auto-wake on limit reset. Local-first ops center for Claude Code — Tauri 2 + React.
Make your AI agent argue visually. Universal skill pack for Excalidraw diagrams across Claude Code, Cursor, Codex, Gemini CLI, and any MCP-compatible agent.
Make the model you already use work more like a frontier model with better planning, persistent context, skills, hooks, failure handling, and verification
Delegate coding work from Claude Code to a local pi agent — MCP tools with mid-run steer/abort, plus per-project discipline hooks.
Multi-model orchestration for Claude Code and Codex, powered by Antigravity with native Windows ConPTY support.
Agent skill that distills AI study chats into a causal-chain learning wiki — mechanical checks, a regression eval gate, evidence-driven self-iteration (RSI). DSH / Codex / Claude Code. | 把 AI 学习问答沉淀为因果链学习 Wiki:机械校验 + 回归门禁 + 证据驱动的自迭代。
GH-300 live demo kit. A real Python app left intentionally immature (bugs, security gaps, low coverage) across 5 structured Copilot modules. Prompt library, Skills, AGENT.md, and a reset script included.
Production-grade prompt templates for Claude Code Routines. 5 recipes using Burn, GitHub, Slack MCP connectors — copy, paste, schedule.
Switch Claude Code between multiple accounts from an animated TUI, with weekly, session and spend usage
Run multiple Claude Code accounts on one machine — per-project switching, live usage limits, and session handoff when you hit a rate limit
Two Claude Skills for behavioral calibration: self-awareness (conversation) and engineering-discipline (coding/data). Reduce hallucination, verbosity, scope creep, and silent data bugs in Claude Code.
Daemon-based hooks system for Claude Code with lazy startup, auto-shutdown, and project-level handler extensibility
Silent context optimizer for Claude Code/Cowork — Native AOT, install-and-forget plugin
Auto-resume Claude Code and Codex across usage-limit resets — iTerm2 + tmux live watchers and a headless wrapper. Dependency-free, macOS/Linux.
Evidence-based audit + cleanup for Claude Code installations. Decide what to keep, delete, or refresh across plugins, skills, and rules — grounded in your actual session history.
Deterministic, auditable agent/skill dispatch for Claude Code. A typed seven-decision scoring kernel runs after the router has read the conversation — post-cognitive routing instead of prose-scanning. Auto-generates its catalog from skill sidecars and agent frontmatter.
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.
Fail-closed Claude Code Stop hook for zero or thin thinking, with optional current-turn stream integration.
Convert any skill, repo, or tool into an install-ready Hermes Agent skill. 20+ formats: Agent Skills, Claude, Codex, Cursor, MCP, plugins, prompt packs.
Qofeno agent skills — real, production-only workflows for Claude Code, Cursor, Antigravity, and other AI agents
Not every coding task needs your best model. Experimental Jev-powered model routing for Claude Code — V3 prototype runs today, V4 routes at the task boundary.
A deterministic context-driven agentic SDLC framework for Cursor and Claude. 41 codified laws and 22 specialist roles to eliminate LLM hallucinations and context drift.
Claude Code plugin with 27 specialist agents, web verification, Red Team review and a final quality gate that blocks weak or unverified answers.
Cross-harness SupervisorLLM plugin for Claude Code, Codex, Cursor and Gemini CLI with native agents, hooks, web verification and final release gates.
🤖 AgentChat 9AI — 9个免费网页AI的Claude Code技能套件(Gemini/ChatGPT/Claude/千问/Kimi/MiniMax/MiMo/DeepSeek/豆包),串行降级+并行编排+深度管道
Parallel multi-agent skills for Claude Code — review, research, investigate, and implement with concurrent specialized agents
Claude Code skill: one session coordinates, spawns and reviews builder sessions across a multi-phase build
A desktop GUI client for the pi agent (pi coding agent). Run an AI agent team — MCP, skills, memory, IM channels. macOS / Windows / Linux, Electron + React.
Event-driven agentic development. Webhook in, coding agent out. No polling loops.
An intentionally small, curated working library of proven operator workflows for AI-assisted software work.
A Claude Code port of pstack: 46 skills and 2 agents for rigorous, parallelizable agent workflows. Orchestration, adversarial review, verification, and 21 engineering principles.
Creative exploration skill for AI coding agents — grounded "what if...?" brainstorming over your codebase, without editing code.
Make your AI agent write in ASD-STE100 Simplified Technical English. Short sentences, no AI slop, plus a linter. Works with Claude Code, Codex, Cursor, and any SKILL.md agent.
Use AI coding agents with LTspice via sim-cli: run circuits, inspect waveforms/logs, and produce replayable simulation artifacts.
Measure your Claude Code skills and rules, and keep only the ones that help. The tool, its evaluation lab and data for the paper arXiv:2610.14216.
The issue-pipeline workflow system as a Claude Code plugin: GitHub Issues as the work-item SSOT, labels as the pipeline, guard hooks, a manager crew, and the skills from build to ship.
Codex skill for coordinating delegated Tasks with one heartbeat per worker, evidence-based progress, and safe shutdown.
A Claude Code skill for fanning work out to independent agent sessions. Completion is decided by a deterministic signal — never by the worker's own report.
Claude Code plugin for second opinions: iterative adversarial debates between Claude and Codex over any subject, from specs, designs, and code to non-technical positions, ending in genuine agreement or a documented dispute.
Claude Code orchestration harness and plugin for long, unattended multi-agent work: validated block plans, worktree-isolated workers, four-gate verification, append-only ledger. Tuned for Claude Opus 5.5.
Four-role adversarial council - Believer, Skeptic, Investor, Judge - as a portable Agent Skill. Rulings carry an evidence-tier justification, a ranked list of what must be true, and the cheapest test that would falsify them.
Install AI-agent governance into any repo: AGENTS.md for every agent, TDD red→green checks, secret scan and architecture guards in a pre-commit hook. Polyglot, monorepo-ready, zero lock-in. AGPL-3.0 with an exception for installed files.
Starter repo for agentic software projects with guardrails built in: AGENTS.md, Claude Code hooks (kill switch, prod-guard, change freeze, audit log), plan-first specs, an agent offer letter, and 46 offline guardrail tests. Companion to the DevOpsDays Cairo 2026 talk Before You Put an AI Agent on Call.
Agent Skill for building, testing and checking Embabel agents (Java or Kotlin, Maven or Gradle)
The missing brain for AI coding agents — verification gates written before the code, Reflexion-style memory that learns from every failure, and self-correction loops with real anchors. One command install for Claude Code; copy-paste prompts for any AI.
Zero-dependency CLI that bootstraps an agentic AI workflow system into any repo. Verbs over Personas.
My personal GNU/Linux dotfiles
Synthetic focus group skill for Claude Code: try a concept on 3 synthetic users before real interviews. Isolated personas react before seeing each other, the moderator probes where they split, and you get 5 interview questions with what answer would reject each assumption.
Desktop screen island & companion for Windows & Linux watching over AI coding agents with Google Gemini API.
Gated Claude Code delivery pipeline: idea to merged PR through PRD, PRP, plan, dev, test, PR, with a deterministic sensor and a scored eval at every stage.
Your Claude quota, live in the terminal status bar — iTerm2, tmux, WezTerm, kitty, starship. Session + weekly usage windows straight from Claude Code's own endpoint.
A Claude Code skill that reviews and redesigns agentic architectures. Be no more agentic than the problem requires: classify every component before allowing an agent, cap autonomy by verification strength, earn complexity through error analysis.
Beautiful, informative status line for Claude Code. Zero dependencies, cross-platform. pip install claude-status
Agent Skill for designing, auditing and tuning relational databases on PostgreSQL and MySQL/InnoDB — schema and key design, index strategy, EXPLAIN-plan diagnosis, locking, and zero-downtime migrations.
Run multiple isolated Claude Code accounts on one machine via a single cc command
Claude Code plugin marketplace by Digital Process Tools. Add with: /plugin marketplace add Digital-Process-Tools/claude-marketplace
VS Code extension that keeps Claude Code's Anthropic prompt cache warm — auto-pings before the 5-minute TTL expires.
🎯 Loadout — profile your project and gear up Claude Code with the right skills, MCP servers & hooks, in one command. A recommender + installer, not a list you read.
Naren's Claude toolkit: skills, mods and MCP servers for Claude Code. One-command install via plugin marketplace.
Claude Code plugin: paces your weekly usage limit - live 'slow down or push' meter, warnings before you burn your quota, /usage command. Zero deps, local-only.
Relocate Claude Code projects without losing sessions, memory, or worktrees.
Move Claude Code sessions to another Claude account or organization: full history, sub-sessions, sync across orgs to keep working past usage limits, and claude.ai chat export to Markdown. Free desktop app for Windows, macOS and Linux.
Claude Code for VS Code 繁體(台灣)/簡體(中國大陸)中文化語言包:自動偵測或手動切換、狀態列快速選單、冪等套用、漏翻偵測。
A marketplace of mods for Claude Code: a step debugger for the agent loop, git account hints, a worktree status line, and more.
Agent Skill for Claude Code that makes the agent search registries, code, papers and official docs before it builds, then write a dated, sourced report you pick from: adopt a candidate, adopt one piece of it, or build what no candidate covers.
Tested agent skills for Claude Code, plus a vetted catalog of third-party skills and plugins. One-command setup.
A port of Spotify's shunt plugin that needs no Portal tenant. Hooks block large file reads and delegate them to a cheap worker model on the CLI you already have: Claude Code, Codex or Gemini CLI. ~90% fewer tokens.
Back up and restore Claude Code, Claude Desktop, and Cowork state on macOS and Windows. MIT-licensed.
9 Claude Code skills recommended by Anthropic team — production-ready skills for code review, CI/CD, debugging, scaffolding, data analysis, infrastructure ops, and more. Based on 'Lessons from Building Claude Code' by Thariq @ Anthropic.
⚡ High-performance linter, cache-layout optimizer, and garbage collector for AI agent instruction files (AGENTS.md, CLAUDE.md, .cursorrules) in Rust.
MasterDeck: the command center for Claude Code. Every session in one window, with your GitHub board, PRs, workflows and costs. macOS and Windows.
Know when your Claude window resets. Phone alerts while your Mac sleeps, a usage dashboard, widgets and mascots for Claude Code.
Unofficial RPM spec for Anthropic's Claude Desktop on Fedora — fixes Wayland/Vulkan crash and Cowork's QEMU firmware paths
Comply — All-in-one compliance skills library for AI coding agents. 12 production-grade skills to get physical products legally to market.
Lloyd Pager 3000 — ESP32 touchscreen status display for Claude Code. Approve/deny tool calls from anywhere with WiFi.
MichiClaude 🐱 — Real-time Claude usage meter for Windows: quota, burn rate & per-project costs
DSH 扩展坞|给 DeepSeek Harness 插上更多接口:快速追加、PTC 派活、中断报警与自用桌宠。全自动派发规格包实测 token 缓存命中 99.6%、约 ¥5.47/亿 token(需严格自动派发 + Ponytail + DeepSeek V4.1-Flash)。
Windows 桌面独立 Claude Code 客户端+VS Code 风格界面,内置 DeepSeek、Kimi、GLM、千问等API预设,填上 key 就开始享受 | Standalone Claude Code desktop app for Windows with a VS Code-style interface and API presets for DeepSeek, Kimi, GLM, Qwen and more: add a key and enjoy
Curated collection of 15 ML/DL skills for AI agents (Claude, Cursor, Cline). Each skill includes instructions, script templates, and reference docs for libraries like PyTorch Lightning, HuggingFace Transformers, scikit-learn, and more.
AI 抬头 | Mid-run self-review nudges for aimless AI coding agents: 检测重复命令/失败循环/零产出并注入提醒,真实本地时钟校准 AI 时间感 (ZCode / DeepSeek Harness plugin)
deepseek app + deepseek harness + termux,极其高效流畅的安卓原生体验,搭配dsh的静态化原生工具与termux执行层,像是增强版的deepseek app
Hand-offs for AI coding agents (Claude Code, Codex, Cursor, any MCP client): a hand-off is refused while a question to that agent is still open. Plain files in git, no server.
Agent-team skills for Claude Code: a debating council, a refactoring-masters panel with a KISS/YAGNI gate, and a browser CEO gate where the human decides.
Multi-agent AI workflow for Claude Code: Claude and Codex plan, implement, cross-review, and merge pull requests.
Local AI agents, one native macOS menu bar. Built with Swift and SwiftUI.
19 Claude Code plugins in one panel. See every move Claude makes: agents, files, plans, tests and risky commands, live.
Open Knowledge Format (OKF) plugin for Claude Code — a skill, auditor agent, and session hook that keep every project's knowledge catalog current.
red-team-challenge: adversarial fact-check / red-team skill for Claude Code (with upgrade dogfood)
Finds which areas of your Claude Code work deserve a set of skills, from the rules you keep restating. Local report, Korean/English.