Human directs. AI delivers.
A true AI agent for pixel-perfect web cloning. Multi-agent architecture built on Claude Agent SDK with 40+ specialized tools. Clones from CSS & structured blocks—not screenshots—enabling perfect reproduction where single-model tools like Cursor/Claude Code fail.
AI-first subjectivity kernel for agents — persistent emotional state, relation dynamics, adaptive reply loops. Zero extra LLM calls.
Idiomatic Kotlin wrapper around com.anthropic:anthropic-java — suspend / Flow / sealed types over the official Anthropic Java SDK.
A Telegram bot with Claude Agent SDK for knowledge-dense lives
One-file setup for Claude Code that drops a complete "plan → tests → code → docs → PR" workflow into any Git repo, powered by Claude AI
Coding Harness - A Human-in-the-Loop Claude Code SDK coding agent with GitLab integration and minimal textual TUI.
Built 100% With CC. Browser light sandboxed self building IDE wrapper for Claude Code
deep research agent for codebase
Define Once. Deploy Anywhere. Govern Automatically. Framework-agnostic platform to build, deploy & govern enterprise AI agents — LangGraph · CrewAI · Claude SDK · OpenAI Agents · Google ADK · AWS · GCP · Azure · K8s · RBAC · A2A · MCP
Claude Code via e-mail based on an addressbook with UNIX style rwx permissions per interaction pair.
Documentation for NanoClaw — a lightweight, secure AI assistant that runs Claude agents in isolated containers with multi-messenger support
Hands-on CCA-F exam prep: Customer Support Resolution Agent demonstrating all 6 architectural patterns (escalation, compliance, tool design, caching, context, handoffs) through 9 Jupyter notebooks with anti-pattern vs correct-pattern contrast. 234 tests. Python 3.13+ / Anthropic SDK.
AI-powered design assistant using Claude's multi-agent architecture. 7 specialized agents for UI/UX reviews, accessibility audits, and design system guidance.
An opinionated Agent Runtime built on top of the OpenAI Agent SDK, inspired by Claude-style agent programming.一个构建在 OpenAI Agent SDK 之上的、具有强约定的 Agent 运行时框架,设计理念源自 Claude 风格的 Agent 编程模型。
AI-Powered LinkedIn Research Platform with Multi-Agent System (Claude SDK) and Browser Automation (MCP)
Claude SDK for Pharo — Anthropic API client
AI-era PM operator skill: GitHub Projects V2 + Notion + Google Workspace orchestration for Claude Code / Cursor / Devin. 31 MCP tools + 9 shell scripts + 536 tests + 12-principle PM Operator Stance (v5.4.0, 200+ sources, MAST 2025 / CALM ICLR 2025 aligned).
Agent Org — AI-native multi-agent system that turns a single idea into a full production-ready software delivery package (PRD, code, architecture, tests, and DevOps) through coordinated autonomous AI teams.
Add a description, image, and links to the claude-sdk topic page so that developers can more easily learn about it.
To associate your repository with the claude-sdk topic, visit your repo's landing page and select "manage topics."