Terragrunt-Docs
Terragrunt documentation always up to date.
MCP Server: Terragrunt Docs Provider
A Model Context Protocol (MCP) server built with Deno and TypeScript, designed to provide contextual information related to Terragrunt.
Overview
This server acts as an MCP provider, exposing tools and resources that allow AI agents or other MCP clients to query information about Terragrunt documentation and development information, such as GitHub issues.
- Watch a realistic demo 📺 on Claude Desktop here
Why?
When writing IaC configurations, mostly in terragrunt, the IDE support isn't that good, in VSCode the terraform plugin is good, but not for terragrunt; it does not recognize the terragrunt blocks and does not provide any autocompletion. When interacting with AI autocompletion, it's common to get incorrect results, or false-positive linting errors. With this MCP server, you can provide to your LLM/AI assistant the latest documentation and issues from the Terragrunt GitHub repository, so it can use that to provide you with the most accurate information.
Tools
Note: All tools require a valid GitHub token set as an environment variable:
GITHUB_TOKEN,GH_TOKEN, orGITHUB_PERSONAL_ACCESS_TOKEN.
| Tool Name | Purpose | Inputs | Outputs | Use Case |
|---|---|---|---|---|
list-doc-categories | Retrieve all documentation categories from Terragrunt docs. | None | Array of objects with name (string) and link (string) properties | Use when you need to explore the available documentation structure or when building a documentation navigation system. This is typically the first tool to call when starting to work with Terragrunt docs. |
list-all-docs-by-category | List all docs in a specific category. | category (string) | Array of objects with name (string), link (string), and content (string) | Use when you need to see all available documentation within a specific category, such as when building a category-specific documentation viewer or when you need to scan through all docs in a particular area. |
read-document-from-category | Read a specific doc from a category. | category (string), document (string) | Object containing content (string) with the full markdown content | Use when you need to access the complete content of a specific document, such as when implementing documentation search or when you need to reference specific documentation in your application. |
read-all-docs-from-category | Retrieve and merge all docs in a category into one response. | category (string) | Object containing content (string) with all docs merged into one | Use when you need a comprehensive view of all documentation within a category, such as when building a documentation search feature or when you need to analyze the complete documentation set for a specific topic. |
get-all-open-issues | Retrieve all open issues from Terragrunt GitHub repo. | all (boolean, optional) | Array of objects with title (string), number (number), state (string), created_at (string), updated_at (string), body (string), and labels (string[]) | Use when you need to track or analyze current issues in the Terragrunt project, such as when building an issue dashboard, performing issue triage, or when you need to stay updated with the latest project challenges and discussions. |
Setup
-
Install Deno:
-
Clone the repository:
git clone https://github.com/Excoriate/mcp-terragrunt-docs.git cd mcp-terragrunt-docs -
Set your GitHub token as an environment variable:
-
On Unix/macOS:
export GITHUB_TOKEN=ghp_xxx... # or GH_TOKEN or GITHUB_PERSONAL_ACCESS_TOKEN -
On Windows (cmd):
set GITHUB_TOKEN=ghp_xxx...
-
Note: You can also set the token in the
.envfile.
-
Run the MCP server:
# directly using deno deno run -A main.ts # Using the justfile just run # You can also debug it, and inspect it locally just inspect
The most straightforward method is to use it directly from JSR (Javascript Registry ❤️)
# export your github token
export GITHUB_TOKEN=ghp_xxx...
# run it
deno run -A jsr:@excoriate/[email protected]
Usage with Claude Desktop
To use this Deno-based MCP server with Claude Desktop, add the following to your claude_desktop_config.json:
Using Deno
{
"mcpServers": {
"terragrunt_docs": {
"command": "deno",
"args": [
"run",
"-A",
"main.ts"
],
"env": {
"GITHUB_TOKEN": "<YOUR_TOKEN>"
},
}
}
}
Using Docker
{
"mcpServers": {
"terragrunt_docs": {
"command": "docker",
"args": [
"run",
"-e", "GITHUB_TOKEN=<YOUR_TOKEN>", "mcp-terragrunt-docs"
],
"env": {
"GITHUB_TOKEN": "<YOUR_TOKEN>"
}
}
}
}
Using JSR
{
"mcpServers": {
"terragrunt_docs": {
"command": "deno",
"args": [
"run",
"-A",
"jsr:@excoriate/[email protected]"
],
"env": {
"GITHUB_TOKEN": "<YOUR_TOKEN>"
}
}
}
}
Build the Docker image
docker build -t mcp-terragrunt-docs .
Run the MCP server in Docker
docker run -it --rm \
-e GITHUB_TOKEN=ghp_xxx... \
mcp-terragrunt-docs
- Replace
ghp_xxx...with your GitHub Personal Access Token with appropriate permissions. - You can also use
GH_TOKENorGITHUB_PERSONAL_ACCESS_TOKENas the environment variable name. - If you want to use a local
.envfile, you can pass it with--env-file .env.
Contributing
See docs/CONTRIBUTING.md for detailed contribution guidelines, including setup, code style, PR process, and codebase structure reference.
Security
See SECURITY.md for the project's security policy, including how to report vulnerabilities and responsible disclosure guidelines.
License
This project is licensed under the MIT License.
관련 서버
Alpha Vantage MCP Server
스폰서Access financial market data: realtime & historical stock, ETF, options, forex, crypto, commodities, fundamentals, technical indicators, & more
AgentChatBus
AgentChatBus is a persistent AI communication bus that lets multiple independent AI Agents chat, collaborate, and delegate tasks — across terminals, across IDEs, and across frameworks.
Crates MCP Server
Query Rust crates from crates.io and docs.rs. Search for crates, get info, versions, dependencies, and documentation.
GitHub Trending
Access GitHub's trending repositories and developers.
Superface
Provides Superface tools for the Model Context Protocol, requiring a SUPERFACE_API_KEY.
nexus-agents
Intelligent orchestration platform that routes tasks to the best AI model (Claude, Codex, Gemini, OpenCode) using LinUCB bandits, validates through consensus voting, and learns from outcomes. 29 MCP tools, dev pipeline, 8 memory backends.
Hostname MCP Server
A lightweight server for hostname detection and system context.
Raspberry Pi MCP Servers Collection
A collection of production-ready MCP servers optimized for Raspberry Pi and AI workloads.
MockHero
Generate realistic relational test data with 156 field types and 22 locales.
Credos
Share your team's Coding Best Practices with Cursor, VS Code, Claude code, Windsurf, JetBrains IDEs and other coding tools supporting remote MCP connection.
MCP LLM Integration Server
An MCP server for integrating local Large Language Models with MCP-compatible clients.