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., "@UAB Research Computing Documentation MCP Server how do I submit a SLURM job on Cheaha?"
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.
An MCP (Model Context Protocol) server that provides AI assistants with access to the University of Alabama at Birmingham's Research Computing documentation.
This MCP server allows AI applications like Claude to search and retrieve information from UAB's Research Computing documentation (https://docs.rc.uab.edu), including:
Cheaha HPC cluster documentation - High-performance computing resources and usage guides
Getting support - Office hours, contact information, and support channels
Software and tools - Available applications, modules, and installation guides
Storage and data management - Data storage systems, quotas, and best practices
Job scheduling (SLURM) - Submitting and managing computational jobs
Contributing - How to improve the documentation
Related MCP server: slurm_MCP
The server provides the following tools:
search_documentationSearch the UAB Research Computing documentation for relevant content.
Parameters:
query (string): Search term or phrase
max_results (integer, optional): Maximum results to return (default: 5, max: 10)
Example:
search_documentation("slurm tutorial")
search_documentation("gpu computing", max_results=10)get_documentation_pageRetrieve the full content of a specific documentation page.
Parameters:
page_path (string): Path to the page (e.g., "getting-started/intro") or full URL
Example:
get_documentation_page("getting-started/intro")
get_documentation_page("https://docs.rc.uab.edu/storage/data-management")get_support_infoGet comprehensive support information including office hours, contact methods, and support channels.
Example:
get_support_info()list_documentation_sectionsList all main sections and categories in the documentation with descriptions.
Example:
list_documentation_sections()get_cheaha_quick_startGet quick start information for accessing and using the Cheaha HPC cluster.
Example:
get_cheaha_quick_start()Python 3.10 or higher
uv package manager (recommended)
Install uv (if not already installed):
macOS/Linux:
curl -LsSf https://astral.sh/uv/install.sh | shWindows:
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"Restart your terminal after installation.
Clone or download this repository:
cd /path/to/uab-rc-docs-mcp-serverCreate virtual environment and install dependencies:
macOS/Linux:
uv venv
source .venv/bin/activate
uv pip install -e .Windows:
uv venv
.venv\Scripts\activate
uv pip install -e .Run the server directly to test it:
uv run uab_docs_server.pyThe server will start and listen for MCP protocol messages on standard input/output.
Install Claude Desktop from https://claude.ai/download
Configure Claude Desktop to use this MCP server:
macOS/Linux:
Edit ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"uab-research-computing": {
"command": "uv",
"args": [
"--directory",
"/ABSOLUTE/PATH/TO/uab-rc-docs-mcp-server",
"run",
"uab_docs_server.py"
]
}
}
}Windows:
Edit %APPDATA%\Claude\claude_desktop_config.json:
{
"mcpServers": {
"uab-research-computing": {
"command": "uv",
"args": [
"--directory",
"C:\\ABSOLUTE\\PATH\\TO\\uab-rc-docs-mcp-server",
"run",
"uab_docs_server.py"
]
}
}
}Important: Replace /ABSOLUTE/PATH/TO/uab-rc-docs-mcp-server with the actual absolute path to this directory.
Restart Claude Desktop to load the new configuration.
Verify the connection in Claude Desktop:
Look for the 🔌 icon in the bottom right
Click it to see connected MCP servers
"uab-research-computing" should appear in the list
This server follows the standard MCP protocol and can be used with any compatible MCP client. Refer to your client's documentation for configuration instructions.
Popular MCP clients include:
Claude Desktop
Continue (VS Code extension)
Zed
And many others (see https://modelcontextprotocol.io/clients)
Once connected, you can ask Claude questions like:
"Search the UAB Research Computing docs for information about GPU computing"
"How do I submit a SLURM job on Cheaha?"
"What storage systems are available at UAB Research Computing?"
"Show me the quick start guide for Cheaha"
"What are the office hours for UAB Research Computing support?"
"Find documentation about using Python on Cheaha"
This MCP server:
Uses FastMCP - A Python framework that simplifies MCP server development
Searches via GitHub API - Searches the documentation repository on GitHub
Fetches content via HTTP - Retrieves documentation pages directly from the website
Follows MCP best practices:
Uses stderr for logging (not stdout, which would corrupt STDIO transport)
Provides clear tool descriptions and parameter documentation
Handles errors gracefully with informative messages
uv run pytestuab-rc-docs-mcp-server/
├── uab_docs_server.py # Main MCP server implementation
├── pyproject.toml # Project configuration and dependencies
├── README.md # This file
└── .venv/ # Virtual environment (created during setup)To add a new tool to the server:
Add a new function decorated with @mcp.tool()
Include a comprehensive docstring describing the tool's purpose
Use type hints for all parameters
Return formatted string results
Handle errors gracefully
Example:
@mcp.tool()
async def my_new_tool(param1: str, param2: int = 10) -> str:
"""
Description of what this tool does.
Args:
param1: Description of param1
param2: Description of param2 (default: 10)
Returns:
Description of what is returned
"""
# Implementation here
return "Result"Check that the path in claude_desktop_config.json is absolute and correct
Ensure uv is in your PATH (run which uv on macOS/Linux or where uv on Windows)
Restart Claude Desktop completely
Check Claude Desktop's logs for errors:
macOS: ~/Library/Logs/Claude/
Windows: %APPDATA%\Claude\logs\
Ensure all dependencies are installed: uv pip install -e .
Test the server directly: uv run uab_docs_server.py
Check for firewall or network restrictions
The search tool queries the GitHub API, which has rate limits:
Unauthenticated: 60 requests per hour
Authenticated: 5,000 requests per hour
For production use, consider adding GitHub API authentication.
UAB Research Computing Documentation: https://docs.rc.uab.edu
Cheaha Access Portal: https://rc.uab.edu
Model Context Protocol: https://modelcontextprotocol.io
FastMCP Documentation: https://github.com/jlowin/fastmcp
To contribute to this MCP server:
Follow the UAB Research Computing documentation contributor guide: https://docs.rc.uab.edu/contributing/contributor_guide/
Submit issues or pull requests to the appropriate repository
Follow Python best practices and maintain compatibility with MCP standards
This MCP server is provided for use with UAB Research Computing resources. Please refer to UAB's policies regarding research computing usage.
For questions about:
UAB Research Computing services: Contact UAB Research Computing support through https://docs.rc.uab.edu/help/support
This MCP server: Submit issues to the repository maintainers
Built by the UAB Research Computing community to make research computing resources more accessible through AI assistance.
Get quick start information for accessing and using the Cheaha HPC cluster.
This tool provides essential information for new users getting started with
the Cheaha high-performance computing cluster at UAB. No parameters required.
Returns:
Quick start guide with essential information for Cheaha access and basic usage
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true, idempotentHint=true, and destructiveHint=false, covering safety and idempotency. The description adds minimal behavioral context beyond this, such as specifying it's for 'new users' and providing 'essential information,' but doesn't disclose additional traits like rate limits, authentication needs, or detailed output behavior. There's no contradiction with annotations.
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 well-structured and concise, with three sentences that efficiently convey purpose, target audience, parameter info, and return value. Each sentence adds clear value without redundancy, and it's front-loaded with the main purpose.
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 low complexity (0 parameters, annotations covering safety, and an output schema present), the description is reasonably complete. It explains what the tool does, who it's for, and what it returns, though it could better differentiate from sibling tools. The output schema handles return values, so the description doesn't need to detail them.
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?
With 0 parameters and 100% schema description coverage, the schema fully documents the input (none required). The description adds value by explicitly stating 'No parameters required,' which reinforces the schema and clarifies usage. This compensates adequately, though no parameter details are needed.
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's purpose: 'Get quick start information for accessing and using the Cheaha HPC cluster.' It specifies the verb ('Get') and resource ('quick start information'), and identifies the target system (Cheaha HPC cluster at UAB). However, it doesn't explicitly differentiate from sibling tools like 'get_documentation_page' or 'get_support_info', which might also provide informational 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 context by stating it's 'for new users getting started,' but it doesn't provide explicit guidance on when to use this tool versus alternatives like 'get_documentation_page' or 'search_documentation.' It mentions 'No parameters required,' which helps with invocation but doesn't clarify selection among siblings.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Retrieve the full content of a specific documentation page.
This tool fetches the complete markdown content of a documentation page from the
UAB Research Computing GitHub repository. Use this after finding a relevant
page with the search tool.
Args:
page_path: The path to the documentation page (e.g., "docs/cheaha/slurm/slurm_tutorial.md" or "cheaha/slurm/slurm_tutorial")
Can be a relative path from the repository root or a GitHub URL
Returns:
The full markdown content of the documentation page
| Name | Required | Description | Default |
|---|---|---|---|
| page_path | Yes |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true, idempotentHint=true, and destructiveHint=false, covering safety and idempotency. The description adds valuable context beyond annotations: it specifies the source ('UAB Research Computing GitHub repository'), content format ('markdown'), and that it fetches 'complete' content, which helps the agent understand behavioral traits like data source and output type.
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?
Well-structured with a clear purpose statement, usage guideline, parameter details, and return value—all in four concise sentences. Each sentence adds value: the first states the action, the second adds source context, the third provides usage guidance, and the fourth and fifth detail parameters and returns efficiently.
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 low complexity (1 parameter, no nested objects), rich annotations (readOnly, idempotent, non-destructive), and presence of an output schema (so returns needn't be explained in description), the description is complete. It covers purpose, usage, parameter semantics, and source context adequately for the agent to select and invoke the tool correctly.
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 provides detailed semantics for the single parameter page_path: explains what it represents ('path to the documentation page'), gives examples (e.g., 'docs/cheaha/slurm/slurm_tutorial.md'), and clarifies format options ('relative path from the repository root or a GitHub URL'), adding significant meaning beyond the bare schema.
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 specific action ('Retrieve the full content'), target resource ('specific documentation page'), and format ('complete markdown content'). It distinguishes from sibling tools by specifying it fetches content after finding pages with the search tool, unlike list_documentation_sections which lists sections or search_documentation which searches.
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?
Explicitly states when to use this tool ('Use this after finding a relevant page with the search tool'), providing clear context and distinguishing it from search_documentation. It implies alternatives by referencing the search tool for initial discovery.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Get information about how to get support from UAB Research Computing.
This tool provides contact information, office hours, and support channels
for UAB Research Computing services. No parameters required.
Returns:
Comprehensive support information including office hours, contact methods,
and links to support resources
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description adds valuable behavioral context beyond what annotations provide. While annotations already indicate readOnlyHint=true, idempotentHint=true, and destructiveHint=false, the description specifies that 'No parameters required' and describes the return content (office hours, contact methods, links to support resources). This provides practical usage information that annotations don't cover.
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 perfectly structured and concise. It starts with the core purpose, adds details about what information it provides, explicitly states the parameter situation, and describes returns. Every sentence earns its place with no wasted words, and information is front-loaded appropriately.
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 (0 parameters), comprehensive annotations, and existence of an output schema, the description is complete. It covers purpose, what information it returns, and the parameter situation. With an output schema present, the description doesn't need to detail return values beyond the high-level summary it provides.
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?
With 0 parameters and 100% schema description coverage, the baseline would be 4. The description explicitly states 'No parameters required', which adds clarity beyond the empty schema. This confirms the tool's simplicity and eliminates any ambiguity about whether parameters might be needed.
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 specific verb 'Get' and resource 'information about how to get support from UAB Research Computing', including what information it provides (contact information, office hours, support channels). It distinguishes from sibling tools like get_documentation_page or search_documentation by focusing specifically on support information rather than general 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?
The description provides clear context for when to use this tool (when needing support information for UAB Research Computing services). However, it doesn't explicitly state when not to use it or name specific alternatives among the sibling tools, though the distinction is implied through the different resource types.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List the main sections and categories available in the UAB Research Computing documentation.
This tool provides an overview of the documentation structure to help users
understand what information is available. No parameters required.
Returns:
A structured list of main documentation sections and their purposes
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true, idempotentHint=true, and destructiveHint=false, so the agent knows this is a safe, repeatable read operation. The description adds context about returning 'a structured list of main documentation sections and their purposes,' which provides useful output information, but it doesn't disclose additional behavioral traits like rate limits or authentication needs. No contradiction with annotations.
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 appropriately sized and front-loaded: the first sentence states the purpose, the second explains the utility, and the third clarifies parameters and returns. Every sentence adds value without redundancy, making it efficient and well-structured.
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 low complexity (0 parameters, read-only, idempotent), the description is complete. It explains the purpose, usage context, parameter requirements, and return value. With annotations covering safety and an output schema likely detailing the structured list, no additional information is needed for effective use.
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?
With 0 parameters and 100% schema description coverage, the schema fully documents the lack of inputs. The description explicitly states 'No parameters required,' which adds clarity and confirms the schema, earning a score above the baseline of 3 for zero-parameter tools.
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's purpose: 'List the main sections and categories available in the UAB Research Computing documentation.' It specifies the verb ('List') and resource ('main sections and categories'), though it doesn't explicitly differentiate from sibling tools like 'get_documentation_page' or 'search_documentation' beyond the scope of listing sections versus retrieving content or searching.
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 by stating it 'provides an overview of the documentation structure to help users understand what information is available,' which suggests it's for initial exploration. However, it doesn't explicitly say when to use this tool versus alternatives like 'search_documentation' or 'get_documentation_page,' leaving the guidance somewhat implicit.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Search the UAB Research Computing documentation for relevant content.
This tool searches through the documentation repository to find pages
that match the search query. Useful for finding information about:
- How to use Cheaha HPC cluster
- Research computing policies and procedures
- Getting support and office hours
- Software and tools available
- Storage and data management
Args:
query: The search term or phrase to look for in the documentation
max_results: Maximum number of results to return (default: 5, max: 10)
Returns:
Formatted search results with titles, URLs, and excerpts
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | ||
| max_results | No |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true, idempotentHint=true, and destructiveHint=false, covering safety and idempotency. The description adds valuable context beyond annotations by specifying the search scope ('through the documentation repository'), result formatting ('titles, URLs, and excerpts'), and default/max values for max_results, which are behavioral details not captured in annotations.
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 well-structured and appropriately sized: it starts with a clear purpose statement, follows with usage context in a bulleted list, and ends with parameter and return details. Every sentence adds value without redundancy, and information is front-loaded for quick understanding.
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 moderate complexity (2 parameters, read-only operation), the description is complete: it covers purpose, usage examples, parameter semantics, and return format. With annotations providing safety hints and an output schema presumably detailing the formatted results, no additional information is needed for effective agent use.
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 schema provides no parameter descriptions. The description compensates by explaining query ('search term or phrase to look for') and max_results ('Maximum number of results to return') with default and max values. However, it doesn't detail query syntax (e.g., wildcards, phrases) or result ordering, leaving some semantic gaps.
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 specific action ('search'), resource ('UAB Research Computing documentation'), and scope ('for relevant content'). It distinguishes from siblings like get_documentation_page (retrieve specific page) and list_documentation_sections (list sections) by emphasizing search functionality across the entire repository.
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 clear context about when to use this tool ('Useful for finding information about...') with specific examples like Cheaha HPC cluster usage and research computing policies. However, it doesn't explicitly state when not to use it or name alternatives among sibling tools (e.g., when to use get_documentation_page instead).
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Each tool has a clearly distinct purpose with no overlap: get_cheaha_quick_start provides cluster-specific onboarding, get_documentation_page retrieves full page content, get_support_info offers contact details, list_documentation_sections shows the documentation structure, and search_documentation enables content discovery. The descriptions reinforce these distinct roles, making tool selection unambiguous for an agent.
All tools follow a consistent verb_noun pattern (get_, list_, search_) with clear, descriptive names that align with their functions. The naming convention is uniform across all five tools, using snake_case consistently without any deviations or mixed styles, making the set predictable and easy to navigate.
With 5 tools, this server is well-scoped for its documentation-focused purpose, covering key user needs: onboarding (get_cheaha_quick_start), navigation (list_documentation_sections, search_documentation), content access (get_documentation_page), and support (get_support_info). Each tool earns its place without redundancy, and the count is neither too thin nor overwhelming for the domain.
The tool surface provides complete coverage for a documentation server, addressing the full user journey: discovering documentation (search, list), accessing content (get page), getting started (quick start), and seeking help (support info). There are no obvious gaps—agents can effectively browse, retrieve, and navigate the documentation without dead ends or missing operations.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Search NVIDIA CUDA documentation and code samples from AI coding agents.
Search and query nTop's knowledge base and engineering guides from AI applications.
The documentation, as a tool your agent can call: 950+ AI-dev guides. Search + fetch tools.
Search, fetch (with provenance), scan, and convert AI instruction files for agents.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/jgordini/rcmcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
