Anthropic agents

Autonomous Project Manager plugin for Claude Code that orchestrates parallel worker sessions via channels to fetch, implement, verify, and merge Linear tasks end-to-end.

Bio-inspired multi-agent orchestration — 4 neuromuscular mechanisms for intelligent LLM routing

Save 60-90% on LLM token costs with intelligent memory compression for multi-agent systems

Autonomous Claude agent fleet in Docker — one orchestrator, N workers, a Redis queue, a north star, and a supervisor that halts polish loops. Rides your Claude Max plan, zero API spend.

Stop burning API tokens. 9 algorithmic engines for real-time context optimization, infinite-loop detection, and smart prompt compression.

An agent composition framework that builds stacks and compiles specialized subagents for Claude Code

Make Claude Code actually work like a senior engineer multi agents, token compression, safety hooks, UI pipeline.

Personalized YouTube digest pipeline — daily summaries filtered through your own lenses via Claude Code routines

Opinionated templates for running Claude Code autonomously on enterprise projects: a CLAUDE.md constitution, a universal Technical Requirements template, and 7 Claude Code skills (architecture, product, code review, OWASP, QA, spec validation, UI/UX) with deterministic PASS/WARNING/BLOCKER checklists. With a filled-in fintech-api example.

Lightweight Telegram relay for Claude Code CLI — remote multi-session AI access from your phone.

Claude Code v2.1.88 source code restored from source map — Anthropic's terminal-based AI assistant (TypeScript/React/Ink)

Discover your journey with Claude — told through its eyes. Together, always.

Iron Man-inspired AI desktop assistant | Claude API · Voice I/O · Live System Monitor · Animated HUD | Python

RecipeScout is an AI-powered iOS recipe discovery app featuring Sage, an intelligent cooking companion powered by Anthropic's Claude Sonnet 4.5, built with SwiftUI and Swift 5. Experience context-aware conversations, time-based meal recommendations, smart meal planning, and auto-generated shopping lists. Showcases SwiftData, MapKit & MVVM design.

Minimal, secure personal AI agent — Claude Agent SDK + Telegram + MCP

Persistent memory for AI agents using git refs. MCP server + Claude Code skill. No vector database, no embeddings, no daemon — branches are scopes, push/fetch is sync.

Study of Claude Code's agent harness internals — architecture analysis, token budget breakdown, offline ops guide, and a Rust reimplementation with 6 crates and MCP support.

Production-ready MCP server template with async PostgreSQL, OpenTelemetry observability, tool-level security, Docker and Kubernetes Helm chart. Zero to deployed in 10 minutes.

Serverless MCP (Model Context Protocol) connector for Google Contacts on AWS Lambda. Per-user OAuth 2.1 authentication with privacy-first design. Enables AI agents & LLMs like Claude to access Google Contacts via People API in real-time. Zero contact data storage. Works with Claude Web & Desktop.

Comprehensive Claude Code agentic scaffold — workflow rules, auto-memory pipeline, FORGE feature loop, hooks, skills, and flavor packs. Extracted from a production SaaS codebase.

Claude Code skills for AI-assisted PRD execution — decompose requirements into atomic REQs, enforce brainstorming before coding, track status per-REQ, and maintain git discipline from first spec to final commit.

Production-grade Claude Code plugin: control your personal Telegram via MTProto (Telethon). 120 typed MCP-tools, 9 resources, 7 prompts, 6 skills, 4-layer PreToolUse safety net, channel-push & permission relay

A Claude skill that rewrites LinkedIn profiles against a target JD with rubric-based refinement. Grounded in real competitor profiles, scored across five dimensions, voice-preserving.

Self-learning memory for AI coding agents — MCP server

OpenStreetMap MCP server providing precision geospatial tools for LLMs via Model Context Protocol. Features geocoding, routing, nearby places, neighborhood analysis, EV charging stations, and more.

Deterministic /compact alternative for Claude Code. Compresses JSONL sessions 4-10% without LLM summarization. Zero hallucinations, full error preservation.

MCP server for International Burch University (IBU) student information system — query grades, attendance, exams, tuition, and notifications via Claude Code or Gemini CLI.

An interactive Human-in-the-Loop (HITL) review tool for Claude. Opens a local UI to 'redline' specifications and plans before code is written.

🐔 An open-source, distributed AI agent farm OS. Raise, train, and trade autonomous agents on your own hardware.

Meta-agent system for Claude Code that creates specialized AI agents through deep domain research and 7-lens quality review

Intent-execution separation for LLM agents: the model emits typed intent (SIF) in a domain vocabulary; a deterministic layer validates, scopes, translates, and executes it. Java/Spring + Vue, with two demo domains (legal, vet).

Automate research paper writing with this multi-agent skill pack for coding agents. Convert unstructured data into LaTeX papers using the PaperOrchestra pipeline.

AI-agentic-first Claude Code framework: bootstrap projects with safety + standards in under 3 minutes; let Steward maintain them autonomously every night.

Cthulhu Mythos-themed agentic harness for Claude Code — 11 Elder God agents, lifecycle hooks, skill system, and multi-tier orchestration

A hands-on tutorial for building a Claude Code-style coding agent with tools, memory, hooks, subagents, and plugins.

VE Harness — run Claude Code for hours without babysitting it. A hosted task board that survives the session, hooks that deny the commit until a review ran, real-time shared memory for your team's agents, and a Docker worker that works your queue overnight.

Long-running delegation architecture for Claude Code: a delegator main session + full-power orchestrator subagents — nested agents that actually delegate, skill-driven spawning, gates, crash-safe resume. Probe-proven physics + cleanroom test harness.

AI Chief of Staff methodology for Claude Code. Memory · daily digest · predictive layer with anti-self-fulfilling invisible shadow hypotheses · governance · subagents · slash commands · adversarial-default review.

⚡ The Claude Code power-rig in 1 prompt: settings.json, CLAUDE.md AGI-proactive, context hooks, memory, 12 plugins + an anti-leak pipeline of secrets to participate in the configuration without risk. 🇫🇷 FR-native.

A Claude Code skill that brings Claude Fable 5's engineering discipline to Claude Opus 4.8 and other models: act decisively, ground progress claims in tool output, verify with fresh context, avoid over-engineering. Distilled from Anthropic's public Fable 5 migration guidance.

🧠 弱智吧.Skill - AI弱智对话生成器 | 解压神器 | 搞笑AI工具 | Claude Code/OpenClaw/WorkBuddy Skill

Self-hosted AI personal assistant for businesses & freelancers — deploy a private, Claude-powered Telegram agent on your own Linux server in minutes. Host-isolated and fully yours.

Claude Code inside Unreal - the MCP layer Epic doesn't ship: real editor Python, viewport perception + scene diffing, one-command lighting moods, normal-aware spatial reasoning, and a lossless UE<->X3D round-trip harness. Honesty enforced by tests; coexists with Epic's Unreal MCP (UE 5.8). 58 tools, MIT.

Stateless reducer that turns a feature description into a 9-file PRD. 17 MCP tools · multi-judge verification with weighted-average + Bayesian consensus · deterministic Hard Output Rules · research-evidence-backed strategy selection · 248 tests · part of the ai-architect ecosystem.

Fight AI context rot: keep Claude Code (and any LLM) sharp in long sessions with one living Checklist instead of lossy /compact summaries. Persistent working memory + context-window HUD + zero-keystroke auto-clear.

An MCP server that gives Claude Desktop structured access to your notes and past papers. Discovers topics, teaches in any style, quizzes you on grounded content, extracts PYQ patterns, and tracks mastery in local SQLite. Built with the official MCP Python SDK + FastMCP.

Connect Claude to live trend data via MCP: Google, YouTube, TikTok, Reddit, Amazon, Wikipedia, news sentiment, App Store, Steam, npm and more. One MCP server, free tier.

MCP server for FastPanel 2 — manage sites, databases, SSL, nginx configs from Claude/Cursor via natural language

Flarepoint | A plaza of small, behavioural Claude Code plugins that make the AI's cognitive state explicit and persistent

A web-based configuration management interface for the Common MCP Gateway. Manage all your downstream MCP servers through an intuitive web interface. #cursorAI #devin #copilot #claude #bmsoft1024

Passive runtime inspector for Flutter. Captures HTTP, logs, exceptions, DB queries; surfaces to CLI and to AI agents (Claude Code) via MCP. Debug-only, zero release overhead.

Enable automatic topic-based memory for Claude Code to maintain context across sessions and data compactions without losing information.

Autonomous Terraform agent takes a prompt or Jira ticket, runs a full validation pipeline, and opens a GitHub PR. No human in the loop.

Multi-agent deep research for Claude Code. Zero API keys. Paste one line, type /research.

Claude Code skill that ships a full production feature while you sleep. No push, you review.

A threat hunter that lives in your terminal with memories in Notion.

Benchmarked Claude Code plugin that makes cheaper Claude models (Haiku, Sonnet, Opus) match Fable 5's output at a fraction of the token cost — auto-routed reasoning levels, verification discipline, closed-loop habit correction, statusline HUD with usage meters, and a portable AGENTS.md export for any AI coding agent.

Pediatric dose-extrapolation agent: a defensible starting dose for a child from adult pharmacokinetics via allometry × organ maturation, with a cited, graded, auditable rationale. Multi-agent (Opus + Sonnet, live PubMed/openFDA). Decision support, not prescribing.

Evidence router & policy engine for coding agents — enforces source choice and proof-of-use through Claude Code hooks, including routes to MCP tools. Local-first; no LLM in the hook loop.

Local-first Python toolkit for parallel Claude and LLM agent orchestration: consensus voting, stigmergy, boss-worker swarms, and benchmarks

A work-in-progress open-source AI network engineer for UniFi, built as a working laboratory for situated multi-agent reliability and structural cascade self-healing.

Build production-ready AI agents in Elixir — structured schemas, tool calling, multi-agent coordination, streaming, and distributed sessions, with first-class Anthropic Claude support.

Agent Discipline Layer (ADL) — keep a coding agent honest and make it prove its work: layered guardrails + a /goal contract Warden verifies against reality and signs. Claude Code today; Grok/Codex CLI/Pi planned. (formerly Claude Layers)

Production-ready agents, your way.

Portable Claude Code skills + agent panel that translate any project or files into any language — codebase i18n, WordPress/gettext, docs — with adversarial QA, per-run domain specialization, and terminology research.

Zero-dependency multi-agent build, automation and review system for Claude Code. 18 agents, 23 skills, a live per-project dashboard, and a /setup-forge onboarding wizard. One command: /forge.

Backup brain for Claude Code. When your Claude Max plan hits its limit or Claude is down, flip one command and your claude -p agents keep running on a local mlx-lm model. Default Claude. Local only when needed.

Route Telegram topics to isolated Claude Code sessions with persistent memory.

A Jarvis-style local second brain — voice-controlled Claude Agent SDK orchestrator with a live neural dashboard, a self-learning markdown knowledge base, and worker agents dispatched into your repos.

Native macOS companion app for Claude Code. Tabbed PTY sessions, a real MCPs/Skills/Marketplace UI, GitHub-style usage analytics, in-place session restart. Signed + notarized Sparkle auto-updates.

A Claude Code skill that builds hierarchical, domain-agnostic agent systems from a single objective, via a central Designator agent. Self-building, cost-aware, model-agnostic.

Develop Your Local Assistant — a local-first AI workspace that turns project work into validated deliverables (briefs, estimates, data models, mockups, test plans, decks), running on a model on your own machine.

Self-grounding autonomous coding pipeline for Claude Code. Agents classify → plan → build (test-seam slices) → review (binary gates) → compound, grounded in a project knowledge base of gotchas & risk checks, held to a Proof Contract: ships only when evidence artifacts prove it. Includes a runnable eval harness that grades every run.

This repository is a starter template for integrating DataDoe MCP with Claude Code CLI in a secure, team-friendly way.

Browser-integrated LinkedIn companion offering intelligent job filtering alongside tailored application refinement.

metahub is building the missing infrastructure for AI tools, in the open — a registry that installs skills, MCP servers, agents and plugins into any harness, free publisher observability, and an eval framework going open source.

The largest index of MCP servers. 18,146+ Model Context Protocol servers searchable at protodex.io

Evidence-based design playbooks for Claude Code — rules synthesized from 407 real product sites: page-type playbooks with pattern frequencies, 9 style guides, per-site hex/type tokens, DESIGN.md generator. Free, local, MIT.

Read-only Model Context Protocol (MCP) server for WordPress + WooCommerce — give Claude and other AI agents access to products, orders, sales reports and posts over the REST API. By wppoland.com

Stops Claude Code from starting a task your usage quota can't finish. Local, zero-dependency CLI that warns or blocks prompts before you hit your 5-hour or weekly rate limit.

Model Context Protocol (MCP) server for the CallRail REST API v3: 57 tools for calls, leads, form submissions, trackers, tags and agency reporting, usable from Claude Code, Claude Desktop, Cursor or any MCP client.

Connect your personal Telegram account to Claude Code using the MTProto protocol instead of the Bot API.

Milestone-driven development workflow for Claude Code and Google Antigravity.

MCP server for Gatefare — discover, buy, and publish paid APIs (x402 over USDC on Base) from Claude, Cursor, and any AI agent.

Autonomous chemical engineering agent: LLM tool-use over RDKit, Antoine thermo, Python+scipy, arxiv literature. 23-problem benchmark across 10 ChemE subdomains.

Ontology-aware RAG framework — RDF/OWL as the source of truth for LLM agents via MCP

🧰 claude-skills-toolkit · ⚡ 15 skills agentic para Claude Code · 🔒 security-audit (12 capas: OSV/KEV/EPSS/SAST/…) · ₿ bitcoin-custody-audit (custodia Bitcoin, 14 etapas) · 🐍 python-lint-guard · 📌 python-deps-pinning · 🏷️ version-bump · 📄 md-to-doc · 📋 yaml-control · 📝 md-lint-fix · 🐳 docker-cleanup · Zero-deps · Cross-platform 🐧🍎🪟

Minimal, secure personal AI agent — Claude Agent SDK + Telegram + MCP

A configurable HTTP/REST MCP server for talking to everything in your homelab — and beyond.

Orchestrate your agents and workflows with Agentsea ADK + BEAM superpowers ⚡️

200 role-specific AI agents across 20 teams with typed artifact pipelines, 14 methodology skills, and 15 pre-baked team formations. The virtual engineering org for Claude Code, Cursor, Codex CLI.

Turn a use case into a grounded agentic system design. An opinionated AI solutions architect as an agent skill.

Optimize development performance by stripping away unnecessary abstractions and focusing on raw code execution.

A /goal engine for Claude Code where completion is EARNED, not announced: acceptance criteria are real shell commands, the Stop hook re-runs every one of them, and only exit 0 ends the session.

Automatic context preservation for Claude Code. Prevents intelligence degradation after auto-compaction using hooks.

🤖 operational-ai-agents · ⚡ 13 agentes para Claude Code · 🧭 repository-evolution · 🏗️ legacy-modernization · 📡 curriculum-evolution · 🌐 portfolio-publication · 🛡️ security-remediation · 🩺 incident-root-cause · 👤 professional-profile · Contrato + gates humanos + evidencia verificable · 39 evals deterministas · Zero-deps · Cross-platform 🐧🍎🪟

海龟汤（情境猜谜/横向思维谜题）创作方法论——四阶段施工框架 | A methodology for crafting turtle soup / lateral thinking puzzles: 4-phase pipeline from embryo to finished puzzle (CC BY-NC 4.0)

The layer a sandbox doesn't give you: a PreToolUse hook that blocks unapproved spending, account creation and fund movement before the call runs, queues them for human review, and proves it actually ran -- including whether your harness fired it for subagent tool calls. Plus a spend ceiling and loop detector.

A Claude skill that rewrites every AI response in grimdark Warhammer 40K Adeptus Mechanicus voice — 5 languages, 300+ vocabulary mappings