Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@chezmoi-documentation-mcp-server search for how to use templates"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Model Context Protocol (MCP) server for the official chezmoi documentation.
This server is inspired by the shape of the AWS documentation MCP server and focuses on the workflows that are most useful for docs-heavy assistants:
Search the official chezmoi documentation
Read a documentation page and convert it to Markdown
Read specific sections from a documentation page
List known documentation pages discovered from the site navigation
Uses the official https://www.chezmoi.io/ site as its source of truth
Converts Material for MkDocs pages into LLM-friendly Markdown
Builds and caches a local in-memory search index by crawling the docs site
Restricts fetched content to the official chezmoi documentation host
Runs over stdio by default
Related MCP server: MCP Docs Server
Python 3.10 or newer
An MCP client such as Claude Desktop, Cursor, or VS Code
If you have uv installed, you can add this server to VS Code from GitHub:
Equivalent CLI command:
code --add-mcp '{"name":"chezmoi-documentation","command":"uvx","args":["--from","git+https://github.com/llamakko/chezmoi-documentation-mcp-server.git","chezmoi-documentation-mcp-server"],"env":{"FASTMCP_LOG_LEVEL":"ERROR"}}'This adds the server to your VS Code user profile.
python3 -m venv .venv
. .venv/bin/activate
pip install -e .{
"mcpServers": {
"chezmoi-documentation": {
"command": "/ABSOLUTE/PATH/TO/chezmoi-documentation-mcp-server/.venv/bin/python",
"args": ["-m", "chezmoi_documentation_mcp_server"],
"env": {
"FASTMCP_LOG_LEVEL": "ERROR"
}
}
}
}If you prefer to install without a local checkout, you can switch the command to uvx or your preferred package runner.
Prefer absolute paths and avoid wrapping the server launch in a shell command. This follows the MCP guidance for reducing local-server execution risk.
search_documentationSearch the official chezmoi documentation. The first search builds a cached index by crawling the site, so it may take longer than later searches.
Parameters:
query: Search terms
limit: Maximum number of results to return
read_documentationFetch a chezmoi documentation page and convert the main content to Markdown.
Parameters:
url: Absolute URL, root-relative path, or documentation path
read_sectionsFetch a page and return only the requested sections.
Parameters:
url: Absolute URL, root-relative path, or documentation path
sections: A list of section names to match against page headings
list_documentation_pagesList documentation pages discovered from the official site.
Parameters:
section: Optional filter applied to URL and title
limit: Maximum number of pages to return
Variable | Description | Default |
| Base URL for docs crawling |
|
| Set to | — |
| Maximum number of pages to crawl for the search index |
|
| Concurrent fetches while building the index |
|
| Cached search index lifetime in seconds |
|
| HTTP timeout per request in seconds |
|
| Maximum redirects to follow after validating each hop |
|
| Maximum HTML response size in bytes |
|
| Maximum Markdown/text characters returned from a page or section |
|
| Maximum search or filter query length |
|
| Maximum requested sections per call |
|
| Maximum section heading match length |
|
| Maximum documentation URL length |
|
| Upper bound for |
|
| Upper bound for |
|
| Maximum tool calls per rate-limit window |
|
| Rate-limit window length in seconds |
|
| Optional custom user agent for HTTP requests. If unset, the server uses its built-in user agent string | built-in |
This server applies a small set of MCP-aligned hardening measures:
Validates tool inputs and rejects oversized URL, query, and section arguments
Restricts fetches to the official chezmoi docs host by default
Validates every redirect hop before following it
Caps response size and returned document length
Sanitizes returned text and Markdown
Uses an in-memory rate limit for tool invocations
Runs over stdio explicitly
Official MCP references used for these choices:
MCP Security Best Practices: https://modelcontextprotocol.io/docs/tutorials/security/security_best_practices
MCP Tools Security Considerations: https://modelcontextprotocol.io/specification/2025-06-18/server/tools
MCP Transports (stdio guidance): https://modelcontextprotocol.io/specification/2025-06-18/basic/transports
Build an MCP Server logging guidance: https://modelcontextprotocol.io/docs/develop/build-server
Run tests:
PYTHONPATH=src python3 -m unittest discover -s testsRun a low-impact live smoke check against the official docs:
PYTHONPATH=src python3 scripts/smoke_check.pyThe smoke check fetches one representative documentation page and verifies that page parsing and section extraction still work. It intentionally avoids a broad crawl so it can be run regularly with minimal load on chezmoi.io.
Validate pinned GitHub Actions with pinact:
brew install pinact
pinact run --checkUpdate pinned GitHub Actions with pinact:
pinact run -uGitHub Actions also includes a scheduled Smoke Check workflow that runs weekly and is available via workflow_dispatch.
Before pushing a release tag, configure the github-release environment in your repository settings.
This project uses Semantic Versioning. Release tags follow the vMAJOR.MINOR.PATCH format, for example v0.1.0.
Recommended settings:
Add required reviewers
Enable Prevent self-review
Restrict deployments to selected tags such as v*
Disable admin bypass if your workflow allows it
Add an environment variable named RELEASE_ENV_READY with the value true
The release workflow is intentionally configured to fail until RELEASE_ENV_READY=true is present in the github-release environment. This prevents accidentally publishing from an unprotected environment.
The workflow also verifies that the tagged commit is reachable from the repository default branch before it builds a release.
For public repositories, the release workflow generates artifact attestations for files in dist/. You can verify them with GitHub CLI:
gh attestation verify dist/* -R OWNER/REPOList documentation pages discovered from the official chezmoi documentation site.
| Name | Required | Description | Default |
|---|---|---|---|
| limit | No | ||
| section | No |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden but only states the action. It does not disclose behaviors like pagination, default limits, how the 'section' parameter affects results, or what 'discovered' means.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, front-loaded sentence with no wasted words. It is appropriately concise but arguably too terse for a tool with optional parameters and sibling alternatives, missing opportunities to add essential context.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the presence of sibling tools and optional parameters, the description is incomplete. It does not clarify when to use this tool instead of search_documentation or read_documentation, nor does it explain parameter semantics. The output schema may cover return values, but the overall context is lacking.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0% and the description offers no explanation of the 'limit' or 'section' parameters. The agent must rely solely on the schema defaults, which lack contextual meaning.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool lists documentation pages from the official chezmoi documentation site, using a specific verb and resource. It distinguishes itself from siblings like search_documentation and read_documentation by focusing on listing all discovered pages.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The use case is implied through the verb 'list' and the resource, suggesting this is the tool for obtaining an overview of available pages. However, it does not explicitly compare with sibling tools or state when not to use them, leaving some ambiguity.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Read a chezmoi documentation page and convert its main content to Markdown.
| Name | Required | Description | Default |
|---|---|---|---|
| url | Yes |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description must carry the full burden of behavioral disclosure. It states the main action (read and convert to Markdown) but does not mention any side effects, prerequisites, or edge cases (e.g., how 'main content' is determined, whether it is a network call, or error behavior). This provides some transparency but lacks depth.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, concise sentence that front-loads the core action ('Read a chezmoi documentation page') and adds the key detail of Markdown conversion. No unnecessary words or repetition.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's simplicity (one parameter, output schema present, no nested objects), the description covers the essential behavior. However, it does not mention the existence of siblings or when to choose this over them, but for a straightforward read operation the description is reasonably complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, so the description must compensate. The single parameter 'url' is given meaning by the description's reference to 'a chezmoi documentation page', indicating the URL points to that page. This adds enough context beyond the bare parameter name, though it could be more explicit about URL formats.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states a specific verb 'read' and resource 'chezmoi documentation page', and adds the conversion detail 'convert its main content to Markdown'. This distinguishes it from siblings like search_documentation (search) and list_documentation_pages (list) by focusing on retrieving a specific page's content.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage when you need a specific documentation page's content as Markdown, but does not explicitly state when to use this tool versus the siblings or provide any exclusions. There is clear context but no explicit comparison or conditions, so it stops short of strong guidance.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Read selected sections from a chezmoi documentation page.
| Name | Required | Description | Default |
|---|---|---|---|
| url | Yes | ||
| sections | Yes |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden. It only states 'Read selected sections' without disclosing what happens if a section is missing, how sections are matched, or any other behavioral details. The description adds little beyond the tool name.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single concise sentence, front-loaded with the key verb and resource. Every word earns its place with no wasted or redundant content.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Although an output schema exists, the description is too sparse to fully guide usage. It does not clarify the expected format of sections, the relationship between url and page content, or any prerequisites. The low parameter semantics and lack of behavioral guidance make this incomplete for a new user.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, so the description must compensate. It mentions 'selected sections' but does not explain what the `url` parameter is for beyond implying it points to a page, nor what the `sections` array should contain (e.g., heading names, slugs). The description adds minimal semantic value.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states a specific action ('Read selected sections') on a resource ('chezmoi documentation page'), distinguishing it from siblings like read_documentation (which likely reads a whole page) and search_documentation.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Usage is implied: the description and sibling names suggest this tool is for reading specific sections rather than entire pages. However, there is no explicit guidance on when to choose this vs read_documentation, or how sections should be identified.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Search the official chezmoi documentation and return ranked matches.
| Name | Required | Description | Default |
|---|---|---|---|
| limit | No | ||
| query | Yes |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description adds that results are 'ranked matches', which is useful behavioral context beyond the tool's name. However, without annotations, it does not disclose any limitations, whether the search covers full text, or any side effects. Still, the read-only nature is implied by 'search', and the output schema covers return details.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, front-loaded sentence that states the action and resource directly. No waste, every word contributes to understanding the core function.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Despite having an output schema, the description lacks usage guidelines and parameter semantics. Given the simplicity of the tool, a complete description should at least note that 'query' is required and 'limit' controls the number of results.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, and the description does not mention the 'query' or 'limit' parameters. The agent gets no semantic guidance on how to construct values, especially the meaning of 'limit' and any constraints on 'query'.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description uses a specific verb 'search' with a clear resource 'official chezmoi documentation' and an outcome 'return ranked matches'. This clearly distinguishes it from sibling tools like read_documentation and list_documentation_pages, which imply different actions.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides no guidance on when to use this tool versus the sibling tools. It does not mention scenarios or alternatives, leaving the agent without contextual selection criteria.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Each tool targets a distinct action: searching, reading a full page, reading specific sections, and listing available pages. There is no meaningful overlap between these operations.
All tool names follow a consistent verb_noun pattern (search_documentation, read_documentation, read_sections, list_documentation_pages), making the set predictable and easy to navigate.
Four tools is well-scoped for a documentation server, covering the essential tasks without unnecessary duplication. Each tool has a clear purpose and earns its place.
The tool surface covers the full workflow for documentation access: discovering pages, searching content, reading entire pages, and reading specific sections. No obvious gaps exist for the stated purpose.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Search Multilo's docs, fetch any public page as clean Markdown, and read the changelog.
Read-only search and Markdown access to liz's public docs, prompts, resources, and an MCP App.
Search, fetch (with provenance), scan, and convert AI instruction files for agents.
DevDocs.io keyless docs index + entry search + content (Angular, MDN, Rust, etc.).
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/llamakko/chezmoi-documentation-mcp-server'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
