Mobi MCP Server
This is a Model Context Protocol server that will allow agentic interfacing with a given instance of Mobi (GitHub).
Prerequisites
- Python 3.10 or higher (required for MCP package)
- A running Mobi instance
Model Context Protocol (MCP) is a standardized communication protocol that enables AI models to interact with external tools and services. When integrated with Mobi, MCP provides several key benefits:
- Seamless interaction between AI models and Mobi's functionality
- Structured data exchange and command execution
- Real-time updates and bidirectional communication
- Enhanced automation capabilities through standardized interfaces
- Platform-agnostic integration with various AI models and services
Installation
- Clone the repository:
- Create a Python virtual environment:Note: If you don't have Python 3.10+, install it first:
- macOS:
brew install python@3.12
- macOS:
- Activate the virtual environment:
- macOS/Linux:
source .venv/bin/activate
- macOS/Linux:
- Install dependencies:
Leveraging the MCP Server
The easiest way to quickly try this is to hook the MCP server into Claude Desktop and then
configure the application to use your MCP server by modifying the
~/Library/Application\ Support/Claude/claude_desktop_config.json
file with a JSON snippet like this:
-- Note: You'll have to adjust the paths as absolute paths are required :)
Claude Desktop leverages the stdio MCP transport, but this server also supports SSE as well (you can run the mobi-mcp
python script with the argument --sse
).
This also assumes that you are running an instance of Mobi as well. If you want to try and run one locally, see instructions HERE.
Configuration
The MCP server requires the following environment variables:
MOBI_BASE_URL
: The base URL of your Mobi instance (e.g.,https://localhost:8443
)MOBI_USERNAME
: Username for Mobi authenticationMOBI_PASSWORD
: Password for Mobi authenticationMOBI_IGNORE_CERT
: Set to"true"
to ignore SSL certificate verification (useful for self-signed certificates)
Repository Structure
The repository is organized to maintain a clean separation between the Mobi API interaction layer and the MCP protocol implementation:
src/mobi.py
- Contains the core Mobi API client functionality, handling authentication, API calls, and data processingsrc/mobi-mcp.py
- The main MCP server implementation that bridges Mobi functionality with the Model Context ProtocolDockerfile
- Enables containerized deployment of the MCP serverMakefile
- Provides convenient commands for environment setup and container managementrequirements.txt
- Lists all Python package dependencies needed for the project
Current Tools
The Mobi MCP Server exposes the following tools for interacting with your Mobi instance:
Search and Discovery Tools
record_search
Search the Mobi catalog for records matching specified criteria.
- Parameters:
offset
(int): Starting index for paginationlimit
(int): Maximum number of records to retrievesearch_text
(str, optional): Text to search for in recordskeywords
(list[str], optional): Keywords to match against recordstypes
(list[str], optional): Record types to filter by. Valid types:http://mobi.com/ontologies/ontology-editor#OntologyRecord
(Ontology/Vocabulary)http://mobi.com/ontologies/shapes-graph-editor#ShapesGraphRecord
(SHACL)http://mobi.com/ontologies/delimited#MappingRecord
(Mappings)http://mobi.com/ontologies/dataset#DatasetRecord
(Datasets)
entity_search
Search the Mobi catalog for records whose metadata contain a specified string.
- Parameters:
search_for
(str): Substring to search for within entities' metadataoffset
(int): Starting point within the result setlimit
(int): Maximum number of entities to returntypes
(list[str], optional): Entity types to filter bykeywords
(list[str], optional): Keywords to filter by
Data Retrieval Tools
fetch_ontology_data
Fetch ontology data for a given ontology record IRI.
- Parameters:
ontology_iri
(str): The IRI of the record containing the ontology data (not the ontology IRI itself)
- Note: Use the record IRI from search results, typically in the format
https://mobi.com/records#<uuid>
get_shapes_graph
Retrieve the shapes graph for a specified record.
- Parameters:
record_id
(str): Unique identifier for the recordbranch_id
(str, optional): Branch identifier within the recordcommit_id
(str, optional): Commit identifier to target
Content Creation Tools
create_ontology_record
Create a new ontology record from RDF data and metadata.
- Parameters:
rdf_string
(str): The RDF data as a stringrdf_format
(str): Format of the RDF data (e.g., "xml", "turtle")title
(str): Title of the ontologydescription
(str): Textual description of the ontologymarkdown_description
(str, optional): Markdown version of the descriptionkeywords
(list[str], optional): Keywords associated with the ontology
- Note: Confirmation with the user is recommended before creating new records
Version Control Tools
create_branch_on_record
Create a new branch on an existing record.
- Parameters:
record_iri
(str): IRI of the recordtitle
(str): Title of the new branchdescription
(str): Description of the new branchcommit_iri
(str): IRI of the commit to use as reference
Usage Notes
- All tools interact with Mobi's git-inspired versioning system
- Record IRIs are typically returned from search operations
- The server follows semantic versioning practices similar to git repositories
- Always verify record and branch IRIs before performing write operations
This server cannot be installed
hybrid server
The server is able to function both locally and remotely, depending on the configuration or use case.
A Model Context Protocol server that enables AI models to interact with Mobi instances, providing structured data exchange and command execution for ontology management, data retrieval, and content creation.
Related MCP Servers
- AsecurityFlicenseAqualityA Model Context Protocol server that enables AI models to interact with SourceSync.ai's knowledge management platform for managing documents, ingesting content from various sources, and performing semantic searches.Last updated -25666
- AsecurityAlicenseAqualityA Model Context Protocol server that enables AI models to interact with both MySQL and MongoDB databases through a standardized interface, supporting comprehensive database operations including queries, schema management, and CRUD operations.Last updated -146JavaScriptMIT License
- AsecurityFlicenseAqualityA Model Context Protocol server that enables AI assistants to interact with the Omi API for retrieving and creating conversations and memories for users.Last updated -42TypeScript
- -securityAlicense-qualityA server that implements the Model Context Protocol, providing a standardized way to connect AI models to different data sources and tools.Last updated -17TypeScriptMIT License