- Home
- MCP servers
- Gigaspec
Gigaspec
- javascript
0
GitHub Stars
javascript
Language
4 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": {
"oleksiitrembach-gigaspec": {
"command": "gigaspec-mcp",
"args": []
}
}
}Gigaspec provides a Model Context Protocol (MCP) server that lets AI IDEs connect to its AI-native specification workflow. This enables you to drive AI-assisted project planning, analysis, and code generation directly from your IDE, with structured prompts and living documentation kept in sync as you develop your app.
How to use
You will connect an MCP client in your AI IDE to one of the available Gigaspec MCP configurations. Once connected, the client can initialize a project, request analysis prompts, generate specification files, check status, and continue with the next development task. Use the MCP client to drive an interactive AI-guided workflow without leaving your editor.
How to install
Prerequisites: ensure you have Node.js version 16 or higher and npm or yarn installed on your system.
Option A — Global MCP server install (recommended) install and enable in your environment.
Option B — Use with npx for a zero-install workflow.
Option C — Run a local development MCP server from source.
Proceed with one of the following methods.
npm install -g gigaspec
# Configure MCP in your IDE for the global server
# Kimi Desktop: set mcpServers.gigaspec.command to gigaspec-mcp
# Claude Desktop: set mcpServers.gigaspec.command to gigaspec-mcp
# If you prefer no install, use npx:
npx gigaspec init --name "MyApp"
# Start the MCP-enabled workflow via CLI for immediate use
gigaspec init --json
# Run a local development MCP server (from source)
node ./bin/mcp-server.js
## Additional setup notes
If you choose the global install route, CRITICAL MCP usage requires you to configure your IDE’s MCP settings to point to the provided command gigaspec-mcp. For local development, you can launch the MCP server directly from the repository with node ./bin/mcp-server.js.
## Security and best practices
Only connect to MCP endpoints you control and keep your project secrets in secure storage. When using the local development option, avoid exposing your MCP server to public networks unless properly secured.
## Available tools
### gigaspec-init
Initialize a new project for AI-assisted development using the MCP workflow.
### gigaspec-analyze
Create an analysis prompt to deeply evaluate project requirements.
### gigaspec-generate
Generate specification files and living documentation based on AI analysis.
### gigaspec-status
Query the current status of the living project state.
### gigaspec-continue
Request the next development task to implement.
### gigaspec-verify
Verify code against AGENT.md to ensure compliance.