plan-execution

A plan engine for dynamic planning and reliable execution of AI agent workflows.

Turn a Markdown execution plan into an isolated, resumable Pi run - durable controller state, one writer, deliberate recovery.

Agent 规划与执行框架｜Plan 即运行时对象｜G4C 五要素 + TAO 步骤级执行 + Replan 自我修正 + 失败回溯 + 信任状态管理

A dedicated Plan Execution Runtime for Pi Coding Agent. Pure, reusable, and source-first engine for scheduling multi-step agent workflows in serial or parallel without business state coupling.

基于 ReAct + Plan-Execute 混合策略的 LLM Agent，支持文本查询和图片识物两种入口，在多电商平台并行比价，并具备语义推荐、引导式购物、RAG 知识增强能力。

Plan execution layer for AI coding assistants. One command — isolate, classify, execute, verify, merge. Works with Claude Code, Cursor, Windsurf, Aider, and any LLM.

Claude Code skills for executing trusted Markdown plans

Make an AI agent's 'done' mean 'here is the command output that proves it.' Deterministic, LLM-free verification for AI coding agents: sealed plans, SHA-256 hash-chained evidence, STRIPS/PDDL reachability, and a fail-closed completion gate.

Plan Executor and Adaptor (PlExA)