Plan Cascade

v1.0.0Coding Agentsstable

AI-powered cascading development framework. Decompose complex projects into parallel executable tasks with auto-generated PRDs, design docs, and multi-agent collaboration (Claude Code, Codex, Aider).

claude-codeclaude-code-plugindeveloper-toolsllm-agentsmcp-server
Share:
81
Stars
0
Downloads
0
Weekly
0/5

What is Plan Cascade?

Plan Cascade is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to ai-powered cascading development framework. decompose complex projects into parallel executable tasks with auto-generated prds, design docs, and multi-agent collaboration (claude code, codex, aider).

AI-powered cascading development framework. Decompose complex projects into parallel executable tasks with auto-generated PRDs, design docs, and multi-agent collaboration (Claude Code, Codex, Aider).

This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • AI-powered cascading development framework. Decompose comple

Use Cases

Decompose complex projects into parallel executable tasks.
Auto-generate PRDs and design docs from requirements.
Orchestrate multi-agent collaboration with Claude and Aider.
Taoidle

Maintainer

LicenseMIT
Languagerust
Versionv1.0.0
UpdatedMay 21, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx plan-cascade

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Plan Cascade

Plan Cascade is a Claude Code plugin and CLI tool that applies a cascading, quality-gated workflow to complex software projects: it decomposes a high-level goal into parallel executable tasks, auto-generates Product Requirements Documents and design documents for each feature, and then orchestrates multiple AI coding agents (Claude Code, Codex, Aider) to implement them concurrently. A built-in quality pipeline validates each task against Definition-of-Ready and Definition-of-Done criteria, detects stub code, and scores the output before merging. Teams building large features benefit from the structured decomposition and the ability to parallelise work across agents.

Prerequisites

  • Claude Code installed and authenticated (for plugin mode)
  • Python 3.10+ and the uv package manager (for CLI mode)
  • API keys for the LLM providers you want to use (Anthropic, OpenAI, DeepSeek, or Ollama for local models)
  • Git repository for the project being developed
1

Install the Claude Code plugin

Install Plan Cascade as a Claude Code plugin with a single command. This is the recommended stable installation method.

claude plugins install plan-cascade
2

Clone the repository for CLI or development use

If you prefer the CLI or want to run tests, clone the repository and install dependencies with uv.

git clone https://github.com/Taoidle/plan-cascade.git
cd plan-cascade
uv run pytest tests/
3

Configure your LLM provider credentials

Set the API key environment variable for your chosen provider. Plan Cascade supports Anthropic, OpenAI, DeepSeek, Ollama, GLM, Qwen, and MiniMax.

export ANTHROPIC_API_KEY=your_key_here
# or
export OPENAI_API_KEY=your_key_here
4

Run the auto decomposition workflow

From within a Claude Code session, invoke the plan-cascade:auto slash command with your high-level requirement. Plan Cascade will generate a PRD, decompose the work, and begin multi-agent execution.

/plan-cascade:auto "Implement user authentication with JWT tokens and refresh token rotation"
5

Review the generated documents and task status

Plan Cascade writes PRD and design documents to the repository during execution. Review the generated docs and monitor the quality gate scores for each task as agents complete their work.

Plan Cascade Examples

Client configuration

Claude Code MCP server configuration for Plan Cascade when using it as a standalone MCP server rather than a plugin.

{
  "mcpServers": {
    "plan-cascade": {
      "command": "npx",
      "args": ["plan-cascade"],
      "env": {
        "ANTHROPIC_API_KEY": "your_anthropic_api_key"
      }
    }
  }
}

Prompts to try

Example slash commands and prompts for Plan Cascade inside Claude Code.

- "/plan-cascade:auto 'Add a payment processing module with Stripe integration'"
- "/plan-cascade:auto 'Refactor the data access layer to use the repository pattern'"
- "Decompose building a REST API for a todo app into parallel tasks and generate a PRD for each feature"
- "Run the quality gate checks on the authentication module implementation"

Troubleshooting Plan Cascade

Plugin install command fails with 'unknown command: plugins'

Ensure you have a recent version of Claude Code that supports the 'plugins' subcommand. Update with 'npm install -g @anthropic-ai/claude-code' and retry.

Quality gates block task completion with 'stub detected'

The stub detector looks for placeholder implementations such as 'pass', 'TODO', or 'raise NotImplementedError'. The assigned agent needs to complete the actual implementation before the task passes the Definition-of-Done check.

Multi-agent execution stalls with no progress

Check that the API keys for all configured providers are valid and have sufficient quota. You can limit execution to a single provider by setting only one API key environment variable.

Frequently Asked Questions about Plan Cascade

What is Plan Cascade?

Plan Cascade is a Model Context Protocol (MCP) server that ai-powered cascading development framework. decompose complex projects into parallel executable tasks with auto-generated prds, design docs, and multi-agent collaboration (claude code, codex, aider). It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Plan Cascade?

Follow the installation instructions on the Plan Cascade GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Plan Cascade?

Plan Cascade works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Plan Cascade free to use?

Yes, Plan Cascade is open source and available under the MIT license. You can use it freely in both personal and commercial projects.

Browse More Coding Agents MCP Servers

Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "plan-cascade": { "command": "npx", "args": ["-y", "plan-cascade"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Plan Cascade?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides