⚡ SYNTHRON The Multi-Agent AI Framework That Actually Works
中西命理 AI Agent · 7 路代理合议 + Planner-Executor 运行时决策 + 主动追问 + 跨系统涌现联结 + 反思镜可验证推理。八字 / 紫微 / 易经 / 风水 / 占星 / 塔罗 / 数字命理 一卷合参。
一个基于 Java / Spring Boot / Spring AI 的 AI Agent 实验平台，支持 Planner-Executor 并行执行、工具调用、文件操作、RAG、记忆系统和多文档分析工作台。
Local Kubernetes-first release gating agent (kind+Helm): planner/executor, policy checks, MLflow integration, traces/metrics.
Production-grade data analytics and ML evaluation framework for interpreting model predictions with standardized metrics, scalable pipelines, and MLOps-aligned design.
A Dockerized, local-first Django coding agent powered by planner–executor architecture and grounded execution validation.
A minimal agent architecture that cleanly separates planning from execution, making reasoning, failures, and tool use fully inspectable.
Local-first dashboard for visualizing ccusage-supported AI coding usage across Codex, Kimi, OpenClaw, and other tools.
Inspired by Claude Code’s opusplan mode: use a stronger model for planning and a cheaper/faster model for execution.
Production-grade autonomous remediation platform on AWS Bedrock. Replaces ReAct loops with a Planner-Executor architecture. 100% pass rate across 11 real-world scenarios. TypeScript · MCP · AgentCore.
A systems-first implementation of agent control: explicit retrieval decisions, planner–executor separation, and auditable memory as core architectural mechanisms.
LangGraph implementation of ReWOO: Reasoning Without Observation framework
Add a description, image, and links to the planner-executor topic page so that developers can more easily learn about it.
To associate your repository with the planner-executor topic, visit your repo's landing page and select "manage topics."