MAC (Multi-Agent CAD): A decoupled multi-agent framework for text-to-CAD generation via constrained test-time compute

Durable workflows for agent teams in TypeScript: named roles, evals that decide the next step, and a person's approval where you want one. Runs Claude Code, Codex and their kin as a team. No server, no database.

Ai-Agent Builder is a production-grade, "backend-less" single-file web application for authoring, configuring, and packaging AI agents. It runs entirely in the browser — no server, no API keys stored, no telemetry — making it invisible to trackers and safe for enterprise environments.

Self-hosted lifecycle platform for software powered by Claude Code. Configurable pipelines, devices you control, no credentials on the server. Apache-2.0.

古籍识别三 Agent 流水线：OCR → 页序重排 → 错别字纠正 | Ancient Chinese text OCR pipeline: 3-agent system for classical books, ¥0.016/page.

Reuse Agent Skills across Cursor, Claude Code, Codex, and CodeBuddy. Build task-specific AI coding agent pipelines for one project, each with an independently selected capability set.

Concurrent and containerised agent pipelines with built-in audit trails

Sequential LLM agent pipelines that lint, tag and enhance markdown notes — hexagonal architecture with a FastAPI backend and web UI

Manifest-driven 21-stage agent pipeline for Python — 5 LLM backends (Anthropic / OpenAI / Google / vLLM / Claude Code CLI), tools, skills, memory, sandboxed CLI runs, MCP. The engine behind Geny. Apache-2.0.

Multi-agent pipeline simulator — planner, researcher, coder, reviewer, synthesizer tracking quality and latency

Your AI dev brigade for Claude Code. Multi-agent sprint pipeline with fan-out review, Bug Council, sprint learnings. Zero dependencies.

Sourced HTML/CSS video pipeline with local TTS and FFmpeg / 本地视频流水线

An open, modular agent pipeline framework for building, orchestrating, and executing AI workflows.

Reads our meetings, Slack, GitHub, email and docs. Finds work that needs doing, and updates the task board. Pauses for a human when needed, and checks the live board first. Offline demo, no API key.

论衡 — 中文学术/深度长文专用流水线。多 Agent 编排 + 三角验证（文献/数据/案例）+ M 门形式合规 + 实战反馈驱动升级。5000+ 字强推

Autonomous design director — a 7-agent Gemini pipeline that plans, composes, critiques and self-corrects production-ready poster layouts from one image plus your intent. Local-first, no CSS required.

four-agent-pipeline agent skill: Specifier -> Coder -> Refactorer -> Architect. Turn a vague requirement into gate-verified code with Gherkin acceptance specs, steel-cage quality gates (coverage, complexity, duplication, mutation score), and automatic failure routing. Zero-dependency Python scaffolder included. MIT.

Agentic Pipeline for course creation. Mechanical verification, diagrams and house style. Highly parallel with guardrails and UI included. VSCode, Mastra and LangGraph Editions.

Hank Agent Team — 多 Agent 协作流水线引擎

Orchestration system for long-running complex agent tasks