llm-workflow 开源项目推荐（119 个）

shuxiachai /academic-commercialization-agent

将研究论文转化为带引用的商业化与技术转移评估的 AI 多智能体工具：TRL、专利格局、市场与风险。使用 CrewAI + FastAPI，并配有可复现的基准测试。

F-R-L /forge-film

多模型DAG驱动的并行AI电影生成——并行加速可随场景独立性扩展;同时生成电影场景，而不是一个个地生成; "把影视生成的执行图从拓扑序变成关键路径最优调度" ;唯一把场景叙事依赖建模为 DAG、以 CPM 算法驱动并行调度的影视生成引擎

huisezhiyin /sdd-riper

面向智能体编码的轻量级 AI 智能体执行框架：让强模型探索，人类通过最小规格、检查点、审批、验证和反向同步进行引导。

ISON-format /ison

🔥 Token-efficient JSON alternative for LLMs & agentic AI — same data, fewer tokens. Python · JS/TS · Rust · Go · C++

escapeboy /agent-fleet-o

开源 AI 智能体编排平台——面向自主多智能体系统的自托管任务控制中心。可视化 DAG 工作流、670+ MCP 工具、人工介入审批。可与 Claude、GPT-4o、Gemini、Ollama、Codex 配合使用。

Junhanliu-dev /espalier-engineering

Train your AI coders the way you'd train a vine — discover your codebase's actual patterns, then encode them as Claude Code rules, skills, agents, hooks, and a guided pipeline so generated code lands inside your conventions on the first try, not the fifth.

wenboxia /liangyi

两仪 Liangyi · 跨厂商多 Agent 产品想法优化工作流

tianji-qingtian /AI-Native

AI-Native 通用开发工作流方法论：4-Phase 框架 + 空项目/已有源码两种场景 + 反模式速查。 让 Claude Code 等 AI agent 在共享契约下并行干活。

Over9rowth /O9-RD-MAO-Governance

Deterministic Software Engineering Governance for Multi-Agent Coding Workflows (CLI, .cursorrules, 4-Artifact Templates & 14-Chapter Guide) | 融合软件工程治理的多智能体编排框架与完全指南

FlowLLM-AI /flowllm

FlowLLM: Build LLM applications with ease.

tianji-qingtian /AI-Native-Code

AI-Native-Code开发工作流方法论：4-Phase 框架 + 空项目/已有源码两种场景 + 反模式速查。 让 Claude Code 等 AI agent 在共享契约下并行干活。

samestrin /qwen-prompts

Supercharge qwen-code with hybrid prompt-chaining, powered by gemini-cli integration. Performance improvements include 72% faster execution, 36-83% token efficiency gains, and 91.7% success rate verified across 5 repository benchmarks.

wenboxia /airadar

AIRadar · 每日定时运行的 AI 行业情报工作流

JinPLu /Teamwork

[DEPRECATED] 已废弃，请改用 https://github.com/JinPLu/teamwork-simplify —— Astra/Fable 等模型已吸收流程脚手架，Skill 已全面重构。

X-T-E-R /assay

Study many. Build your own. A CLI workbench for evaluating external frameworks and absorbing what proves out into your own systems — with evidence and decisions on record.

samestrin /llm-env

Force any OpenAI-compatible or Anthropic API-compatible tool (Claude Code, Qwen Code, Aider, Fabric, Interpreter) to use Gemini, Groq, Cerebras, or Ollama. Pure Bash. Zero dependencies.

freshdigital-it /claude-execution-harness

Claude Code autopilot with self-healing memory: one command → build → test → localhost. Trajectory tracking, learned routing, and stop-hook enforcement across runs.

ryan-nguyen-01 /maestro

Multi-agent AI workflow platform — coordinator-driven, 11 agents, 227 skills, 15 rules.

ningbo00 /token-saver-loop

Portable reviewer/worker workflow kit for AI coding agents. Reduce premium model context waste by moving execution, retries, and evidence collection to bounded worker rounds.

eddlev /vm4ai-air-kit

AI work, carried forward - a prompt-based framework for focused execution, structured handoff and continuity across sessions and platforms.