Strategize
Define goals, boundaries, risks, and acceptance criteria first.
AI Engineering Command · Local-First
One command spawns order; a hundred tools sheathed, all evidence archived. LingXiao weaves goals, tasks, tools, and verification into a traceable local engineering fabric: define the stance, divide the blades, then close the scroll.
npm install -g @lingxiao-office/lingxiao-coding git clone https://github.com/hexian2001/lingxiao-coding.git && cd lingxiao-coding && npm install && npm run build && npm link 

Product Positioning
LingXiao doesn't chase one-off sparks of brilliance. It breaks complex delivery into executable, observable, and recoverable engineering processes. From goal to specialist分工, to evidence-based acceptance — every step leaves a trail.
Six Weapons Form the Array
Define goals, boundaries, risks, and acceptance criteria first.
Order tasks via DAG — parallel where possible, blocking where needed.
Assign specialists: research, coding, testing, review — each in its place.
Build, test, browser, logs, and contracts all leave verifiable traces.
Blackboard, memory, Git, and sessions preserve every decision.
Eternal Goals advance continuously — complete and close, or surface blockers.
Architecture
Use Cases
API contracts, frontend/backend implementation, browser verification, and docs — all in sync.
Tech debt, test gaps, migration plans, and fix tasks continuously patrolled.
Office, PDF, MCP, Workflow, and local toolchains — unified orchestration.
Proof of the Blade
Both projects were autonomously built by LingXiao AI — from architecture to deployment, fully AI-driven
Open Ecosystem
LingXiao's backend has been split from a single CLI into independently publishable packages: the orchestration core and the Web service layer are decoupled, so you can embed just what you need without the full CLI environment.
Full CLI tool + WebUI + TUI. A ready-to-run AI coding assistant terminal with Leader orchestration, task DAG, and blackboard knowledge graph built in.
npm i -g @lingxiao-office/lingxiao-coding A reusable backend engine SDK: orchestration kernel, Agent runtime, LLM clients, tool system, memory system. No Web layer dependency — embed it directly in any Node.js project.
npm i @lingxiao-office/sdk An HTTP / SSE / WebSocket service layer built on the SDK, letting any product line consume the backend engine over Web protocols — ready to serve out of the box.
npm i @lingxiao-office/web-api LingXiao v1.13.1