CastPlan

Vibe Cording Doc Automation MCP

typescript

GitHub Stars

typescript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "ghostseller-castplan_mcp": {
      "command": "castplan-mcp",
      "args": [
        "start"
      ],
      "env": {
        "CASTPLAN_LOCAL_ONLY": "true",
        "CASTPLAN_WATCH_MODE": "true",
        "CASTPLAN_AI_PROVIDER": "local",
        "CASTPLAN_ENABLE_DOCS": "true",
        "CASTPLAN_PROJECT_ROOT": "/path/to/your/project",
        "CASTPLAN_CACHE_ENABLED": "true",
        "CASTPLAN_MAX_CONCURRENT": "3",
        "CASTPLAN_ENABLE_PLANNING": "true",
        "CASTPLAN_ENABLE_WORKFLOWS": "true",
        "CASTPLAN_ENABLE_AI_ANALYSIS": "true"
      }
    }
  }
}

CastPlan MCP helps your AI assistants remember and reason about your entire project. By maintaining architecture, history, and team decisions locally, it delivers context-aware coding help, automatic documentation updates, and streamlined workflows without exposing your code outside your machine.

How to use

You integrate CastPlan MCP with your AI assistant to give it persistent, project-specific memory. Once configured, your AI can analyze your project structure, base decisions on your past work, and apply consistent patterns across features. Use it to plan work, keep documentation in sync, and automate routine development tasks. Start by initializing the MCP locally, then connect your preferred AI assistant and begin asking for context-aware guidance.

How to install

# Install the CastPlan MCP CLI locally
npm install -g @castplan/automation-mcp
castplan-mcp init

# If you prefer an alternative setup method, you can also run from GitHub:
# No installation needed - runs directly from GitHub!
uvx --from git+https://github.com/Ghostseller/CastPlan_mcp.git castplan-mcp init

Configuration and how it runs

CastPlan MCP can be configured via a standard MCP configuration that defines how to start the local server and which environment variables to use. The configuration below shows a local stdio server that you run on your machine to provide context to your AI assistants.

{
  "mcpServers": {
    "castplan_mcp": {
      "command": "castplan-mcp",
      "args": ["start"],
      "env": {
        "CASTPLAN_PROJECT_ROOT": "/path/to/your/project"
      }
    }
  }
}

Security and privacy notes

CastPlan MCP runs locally by default, and your code remains on your machine unless you explicitly enable external AI services. You can configure local-only operation by setting environment variables accordingly and review all data paths before enabling any remote processing.

Troubleshooting tips

If you encounter issues starting the local MCP, verify the CLI is installed and that your project root path exists. Ensure the start command matches the configuration and that required environment variables are set. Check that your AI assistant can access the local runtime without network restrictions.

Available tools

bmad_parse_specification

Turn requirements into actionable development tasks for project planning.

bmad_update_task_status

Track progress of development tasks and update their status.

docs_reference

Find relevant documentation references for current work.

docs_update

Keep documentation in sync with code changes.

docs_search

Search project documentation for quick references.

analyze_document_quality

AI-powered analysis to assess and improve documentation quality.

hooks_setup_git

Automate integration with Git workflows.

hooks_start_watching

Monitor files for changes to trigger updates.

track_document_work

Connect code changes to documentation updates.

castplan://status

Check the current health status of the MCP setup.

castplan://tasks

View active project tasks managed by CastPlan.

castplan://documentation/history

Track changes to documentation over time.

Built by

VeilStrat

AI signals for GTM teams