@crazyrabbitLTC
基本情報
A MCP server to help with Vibecoding
Begin the feature clarification process
Answer clarification questions about a feature
Generate a Product Requirements Document and implementation plan
Create a development phase for a feature
Add tasks to a development phase
Update the status of a phase
Update the completion status of a task
Get guidance on what to do next
Get the path of a generated document
Save a document to a specific location
Vibe-Coder MCP Server is a Model Context Protocol server that implements a structured development workflow for LLM-based coding. It helps LLMs build features in an organized, clean, and safe manner by providing clarification, documentation, task tracking, and document storage.
Install dependencies with npm install, build the server with npm run build, then configure it in your MCP client’s config file (e.g., claude_desktop_config.json on MacOS or Windows). Use the provided tools and prompts to guide feature development.
Install dependencies with `npm install`, build with `npm run build`, then add the server configuration to your MCP client’s JSON config file (e.g., `claude_desktop_config.json`). The server runs over stdio.
Documents are saved automatically to `documents/{featureId}/` by default, with filenames like `prd.md` and `implementation-plan.md`. You can also use the `save_document` tool to write to a custom path, and `get_document_path` to retrieve the current path and save status.
Tools include `start_feature_clarification`, `provide_clarification`, `generate_prd`, `create_phase`, `add_task`, `update_phase_status`, `update_task_status`, `get_next_phase_action`, `save_document`, and `get_document_path`.
The workflow follows five steps: 1) feature clarification, 2) documentation (PRD and implementation plan), 3) phased development with tasks, 4) progress tracking, and 5) completion verification.
Use the MCP Inspector by running `npm run inspector`, which provides a URL to debugging tools in your browser. The server communicates over stdio, so the Inspector is the recommended debugging approach.
See and control a real Windows PC you own — from any MCP client, locally or remotely. Observe (UIA + screenshots), click/type/drag/scroll, launch apps, owner Live View. BYOH: your machine, your key.
A search tool helps dev to solve the naming things problem.
A curated list of references for MLOps
不止于 CSS
Handl speaks the Model Context Protocol. Your whole billing operation — projects, invoices, clients, reports, approvals — as tools any AI assistant can call, inside your autonomy rules.
コメント