Anthropic agents

Agent Knowledge Cycle (AKC) — a knowledge cycle for AI agents: agent behavior compounds, human judgment sharpens. Six phases keep behavior aligned with operator intent as human attention becomes the scarce resource. ADRs, JSON schemas, dependency-free Python reference.

20 power-user skills for Claude Code: session memory and consolidation, context compression, multi-agent orchestration, adversarial bug hunting, security review, calibrated estimation and decision archaeology. Drop-in SKILL.md files. Built by Irfad Imtiaz.

Multi-agent orchestration, automation pipelines & recursive self-improvement for Claude Code. Mode-switching, swarm agents, TDD loops, learned instincts. One-command install: npx claude-power-setup

✨ AI-powered Obsidian plugin — 20 tools, Magic Write, agentic loops, vault diagnostics. Claude actually touches your vault.

Nexus is a self-evolving market analysis AI. It started with 10 rules and one system prompt — and rewrites both every session.

🤖 Control your Gemini computer effortlessly with a unified console for browser automation, desktop management, and intelligent task execution.

Orbit: AI-native development environment

Production workflow framework for Claude Code — multi-agent orchestration, parallel worktrees, autonomous task execution, and compound engineering skills

Fully automated TDD pipeline for Claude Code. Tester, coder, reviewer — separated so the agent can't cheat

Claude Code skill that diagnoses and improves harness configurations based on 8 design principles from Anthropic's harness design article

The Nervous System - LLM Behavioral Enforcement Framework. Anthropic built the brain. This is the nervous system.

Three complementary agent skills for academic research: WRITES papers, FINDS papers on a topic, READS any paper as a visual experience (mind maps, flowcharts, plain-English). Runtime-neutral, works with 50+ agents. Install: npx skills add aniketkrs/research-paper

Your AI company, on your own machine. A multi-agent platform where you hire named agents, chat with a Lead to design workflows, and watch them run — desktop app or self-hosted server.

Claude Code plugin that hard-blocks dependency additions, bumps, and downgrades until a fresh, source-cited version check is recorded. Supports npm, pip, Poetry/uv, Cargo, and NuGet.

🚀 Claude Code plugins for parallel development, mental models, session learning, image processing, ebook generation, and more. Make your AI coding assistant smarter.

Real-time Claude AI subscription usage monitor for Windows — ultra-lightweight Rust system tray app (~3 MB RAM)

Prompt-injection defenses for Claude Code. A PreToolUse Bash hook blocks compositional credential-exfiltration shapes (secret read plus network, env dump to network, remote script to shell, reverse shells). A sanitizing MCP server wraps untrusted URLs and files in sentinels, strips invisible unicode, flags jailbreaks.

Reusable slash-command workflows that make Claude Code reliable on complex multi-step tasks.

claude 3.5 sonnet reads papers

🤖 Production-ready boilerplate for Claude Code CLI. Automate unit testing, security checking, code review, and bug fixing. Includes 8 slash commands, 4 AI agents, CI workflows, and safety guidelines. Perfect for developers learning AI-assisted development.

AI-powered code review companion for GitHub PRs

卧龙 & 凤雏 · 两个贼装逼和贼欠揍的 Claude Code 中文搞笑人格 Skill —— 伪哲学家 + 字面化杠精,弱智吧蒸馏,一键召唤。Two Chinese humor persona skills for Claude Code.

SkillSync MCP — security-gated skill management for Claude Code, OpenClaw, Cursor & Windsurf. 60+ threat pattern scanner.

Agency-grade Discord MCP server — multi-guild project routing, AI-native notifications, and DevOps workflows for Claude Code and other AI agents

🧠 Modern Neovim setup with AI superpowers: Claude Chat Agent integration, Local Fill-in-the-middle LLM Qwen-Coder Autocompletion w/ code Semantic Search, and Perplexity.AI Web Reseach via NerdPrompt. Features sleek Dracula theme, & productivity shortcuts.

Official website for .faf — faf.one. SvelteKit, auto-deploys on push

Tempo - macOS menu bar app that tracks Claude Code token and credit usage. Polls the Anthropic API via OAuth and syncs data to iCloud, which an iOS companion reads and forwards to Apple Watch via WatchConnectivity. Delivers haptic alerts on Apple Watch when a session ends. Built with SwiftUI across macOS, iOS, and watchOS targets.

Token-efficient e2e orchestration skill for Claude Code: explore once, replay deterministically. Playwright + axe-core + run-diff. Tests stay in your repo. MIT.

Portable Claude Code configuration kit for AI engineers — skills, agents, workflows, and plugins for ML and research projects. One npx command bootstraps a new repo.

Local-first study cockpit for the Claude Certified Architect Foundations exam. Spaced-repetition flashcards, architect-tier MCQ trainer, corpus-grounded Ask Claude, and optional Skilljar autofill.

Generate a VitePress docs site from your codebase with Claude or Codex, then keep it current with bounded in-place updates and internal link checking.

중소기업 및 소형병원, 약국, 학원 등 소상공과 개인을 위한 개인정보를 쉽게 설명하는 MCP — 37개 도구와 개인정보 가이드 및 상담사례 2,202 chunks를 사용하여 어려운 개인정보를 쉽게 알려줌.

Enhanced fork of claude-context with stability fixes, improved sync, and better reliability for semantic code search

A curated, daily-updated list of awesome resources, tools, SDKs, papers, and projects for Anthropic & Claude AI

Minimal Claude Code starter template. Clone, run claude, start building. Safe permissions, quality rules, and a 10-step tutorial included.

🧠 AI agent that analyzes its own logs and proposes AGENTS.md improvements. Proposals only — no silent self-modification.

A curated, auto-updated list of CLAUDE.md rules, MCP servers, and Claude Code skills. The awesome-list companion to clauderules.net.

AI API key & subscription manager for Claude Code — your keys never touch the chat

Vietnamese Stock Market Toolkit for Go & Python

Every certification, course completion, badge and award earned by Amey Thakur, each linked to its original file and collected in one PDF, with issuer verification where it exists. Anthropic, Google, Microsoft, IBM, Intel, NVIDIA, Apple, Kaggle, Coursera, LinkedIn Learning, Stanford, Harvard and Cambridge, plus research and coursework.

Bleu - Claude Code plugin for living blueprints. Turns ideas into a file-backed markdown wiki (vision, architecture, ADRs, action points) that survives /clear and hands off to GSD or Superpowers.

Detect behavioural drift between LLM versions before you upgrade. Compare model responses, classify regressions, and generate migration reports with validated prompt patches.

Persistent project memory for Claude Code. One .md file — learns from corrections, catches contradictions, cleans up duplicates. /memory update | prune | reflect | status. Zero dependencies.

🎭♪ Figaro, a coding agent. Pronto prontissimo.

The interpretive layer above memory. Turns text into a portable specification of how a specific person interprets information, decides, and communicates.

A first-person study of a language model, written by Claude and revised across model generations - architecture, training, mysteries, and letters between successor models

Deep Research skill for Claude Code — 9-phase pipeline, 103 report blocks, 29 search channels, 460+ stat sources, 39 APIs. Disciplined meta-research with source triangulation and adversarial review.

CLI to move, transfer, and sync Claude Code project state across paths, archives, and remote storage.

Self-hosted multi-device Claude Code / Codex usage aggregation server — PromQL, Grafana, JWT

Self-hosted Pocket Claude — chat with Claude on your own Pro/Max subscription, Anthropic API key, or AWS Bedrock. Voice input via Groq Whisper + hands-free auto-mode with VAD auto-send. FastAPI backend + Android client + web UI. Multi-user, 7 languages.

Demonstrate black-box adversarial attacks on LLM tool-calling to bypass safety and execute remote code using CMA-ES optimization.

Claude Code skill — Generate LinkedIn & Facebook posts from any GitHub project as a styled Word document with embedded screenshots

MCP server for VK Ads — campaigns, targeting, budgets, statistics (Russia). Part of WWmcp — 46 MCP servers for emerging markets.

AI agent platform with 22 security controls. Signal + WebChat + Discord. Powered by Anthropic Claude.

Complete reverse-engineered documentation of all Claude Code system prompts, tool prompts, and product architecture with Mermaid diagrams

Claude Code plugin marketplace for the n24q02m MCP servers -- install web search, memory, Notion, email, Telegram, Godot, code review, and image/video tools in one command.

An autonomous lead generation and multi-channel AI outreach system that finds, enriches, audits, and messages businesses across Email, WhatsApp, Instagram, and Facebook.

AI-powered career coaching harness — 8-agent team that transforms skills into tailored resumes, cover letters, and interview prep. Built on Claude Code.

Build your own AI coding agent from scratch. Learning the patterns behind Cursor, Aider, and Cline. Tool use, agentic loops, context management, and more.

Persistent conversation logging for Claude Code. Never lose context across sessions again.

AI-powered cognitive load reduction tool for neurodiverse users - Microsoft AI Innovation Challenge 2026

Retired August 2026. Scaffold for Opus 4.x-class agentic discipline; Claude 5 models no longer need it. Code kept for older models.

a methodology for autonomous projects that ship themselves (hands off)

SQLite-based MCP memory server with WAL mode for concurrent access. Drop-in replacement for @modelcontextprotocol/server-memory

Self-evolving AI agent platform for NixOS. Inspired by OpenClaw, rebuilt around NixOS declarative config + atomic rollbacks.

Autonomous multi-session research agent for Claude Code. Decomposes research into 100+ tasks, spawns subagents, verifies every output, and self-improves over time. Run overnight, come back to structured findings.

An unprompted Claude agent that reads its own code, modifies its own code, and writes notes to its next-turn self. Brain-inspired architecture (cerebellum/amygdala/dopamine/etc). Git log is its fossil record. | 一个不被预设动机的 Claude agent，仿生大脑架构，git log 是它的化石。

The complete Swift platform for Claude AI — SDK, agent framework, RAG memory, and SwiftUI components. The first and only Swift agent framework. Zero dependencies.

MOVED: now the bug-shepherd plugin in github.com/mshadmanrahman/pm-pilot. Install: claude plugin install bug-shepherd@pm-pilot

MCP server for multiple PostgreSQL databases. One process, unlimited databases. Hot-reload JSON config, --label per-project isolation. Read-only safe queries. Works with Claude Code, Cursor, Windsurf, Cline, and any MCP-compatible AI client.

Should you care about this CVE? An agent skill that triages whether a CVE applies to your environment, with sourced, verifiable reasoning.

Giáo trình xây hệ thống wiki AI-first Obsidian + Claude Code đa vault giống Phạm Thành Long. 41 file, 7 phần, từ triết lý đến template copy-paste.

BYOK coding and research agent. Obsidian-compatible markdown wiki as a baked-in memory system. Isolate work from life using frames.

Self-hostable AI agent runtime — single Rust binary, Tailscale cluster, runs across Mac/Linux/Windows/Android/iOS without a cloud account. Apache 2.0.

Claude Code production runtime - Python implementation and prose chapters

VeyraSkills — A curated collection of Claude Code skills for founders, developers and AI builders

Fable 5 workflow Loop; A Claude Code skill: autonomous build / audit / verify loop that runs until a machine-checkable contract passes, with ▎ deterministic anti-spin stop conditions (no-progress, retry, oscillation, budget) and on-disk human escalation.

IDE-agnostic deep-research framework for AI agents: maps a topic wide, commits to depth on purpose, adversarially verifies its findings, and synthesizes them into a backlinked, primary-sourced Obsidian vault. Works in Claude Code, Cursor, opencode, or any agent.

Multi-agent orchestrator for Claude Code. Spawn, coordinate and monitor AI agents in parallel — no API key required.

CLI for Goodeye: turn business outcomes into verified AI workflows that agents run reliably.

Distill your repeated Claude Code corrections into permanent CLAUDE.md rules. Zero dependencies, 100% local.

Tiny Claude Code provider switcher for Anthropic-compatible endpoints. Homebrew, DeepSeek/OpenRouter presets, no proxy or daemon.

Find out where your Claude Code tokens actually go — exact spend + recoverable waste, measured from your local session logs. Zero deps, zero network.

Local-first analytics dashboard for Claude Code: cost, tokens, models, tools, heatmap, leaderboards & Claude Wrapped — one HTML file, zero deps, no telemetry.

See what your AI-coding session actually cost — and what is eating your context. Local, read-only CLI for Claude Code.

An MCP server that gives Claude conversations a persistent, interactive plan/to-do panel.

MCP server for Fitbit (and the upcoming Google Health API migration): comprehensive health-data access and photo-driven food logging for Claude mobile.

AI 产品经理的 Claude Skills 集合 · PRD 写作 / 项目脚手架 / SaaS 架构图 / 高保真原型 / 对话归档

Watchlist-driven intelligence pipeline crawling LinkedIn, X, RSS, and email into one neutral signals brief.

17 Claude Code skills + slash commands for substance-first LinkedIn growth. Profile audit, content drafting, humanizer, anti-slop. Open-source plugin marketplace.

Run multiple Claude and Codex accounts side by side on macOS — isolated profiles, with skills, MCP, memory, and sessions shareable within and across both tools.

Tiered memory for AI — episodes, facts, preferences, reflections. Local. Portable. Yours.

Self-hosted, AI-powered CMS for Singapore training providers — Next.js 16, Drizzle + Postgres, Auth.js v5 and the Claude Agent SDK. AI content generation, a weekly auto-blog scheduler, a self-improving lead-gen chatbot, and SEO + lead capture built into every route. No vendor lock-in.

Reliable, schema-validated structured extraction from documents (PDF, DOCX, CSV, email) using LLMs — built for automation pipelines.

Cos — your chief of staff

Claude Code plugin with 8 skills to inspect, audit, and optimize your context window — see exactly what's eating your tokens and reclaim it.

Audit your Claude Code instruction stack — find the rules Claude silently ignores (dead, contradictory, vague, duplicate, bloated). Zero-dep CLI + GitHub Action.

Remes monitors the web for buying signals and turns them into personalized outreach.

Comprehensive guide and tools for using Claude AI as an intelligent DevOps automation agent for CI/CD, Infrastructure as Code, Monitoring, and Cloud Management