Rasdaman MCP Server
OfficialClick on "Deploy 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., "@Rasdaman MCP Server Show me the dimensions and metadata for the 'Sentinel2_10m' coverage"
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.
Rasdaman MCP Server
This tool enables users to interact with rasdaman in a natural language context. By exposing rasdaman functionality as tools via the MCP protocol, an LLM can query the database to answer questions like:
"What datacubes are available?"
"What are the dimensions of the 'Sentinel2_10m' coverage?"
"Create an NDVI image for June 12, 2025."
The MCP server translates these tool calls into actual WCS/WCPS queries that rasdaman can understand and then returns the results to the LLM.
Installation
pip install rasdaman-mcpRelated MCP server: mcp-geo-server
Usage
First the connection from the MCP server to rasdaman needs to be configured, either through environment variables:
RASDAMAN_URL: URL for the rasdaman serverRASDAMAN_USERNAME: Username for authenticationRASDAMAN_PASSWORD: Password for authentication
or command-line arguments to the rasdaman-mcp tool:
--rasdaman-url: URL for the rasdaman server (defaultRASDAMAN_URLenv variable orhttp://localhost:8080/rasdaman/ows).--username: Username for authentication (defaultRASDAMAN_USERNAMEenv variable orrasguest).--password: Sets the password for authentication (defaultRASDAMAN_PASSWORDenv variable orrasguest).
Then the MCP is ready to be used with an AI agent tool, in one of two modes: stdio (default) or http.
stdio Mode
Used for direct integration with clients that take over managing the server process and communicate with it through standard input/output.
Generally in your AI tool you need to specify the command to run rasdaman-mcp:
rasdaman-mcp --username rasguest --password rasguest --rasdaman-url "..."Example for enabling it in gemini-cli:
gemini mcp add rasdaman-mcp "rasdaman-mcp --username rasguest --password rasguest"Benefits:
Simplicity: No need to manage a separate server process or ports.
Seamless Integration: Tools are transparently made available to the LLM within the client environment.
http Mode
This mode starts a standalone Web server listening on a specified host/port, e.g:
rasdaman-mcp --transport http --host 127.0.0.1 --port 8000 --rasdaman-url "..."The MCP server URL to be configured in your AI agent would be http://127.0.0.1:8000/mcp with transport streamable-http.
For example, for Mistral Vibe extend the config.toml with a section like this:
[[mcp_servers]]
name = "rasdaman-mcp"
transport = "streamable-http"
url = "http://127.0.0.1:8000/mcp/"Benefits:
Scalability: The MCP server can be containerized (e.g., with Docker) and deployed as a separate microservice.
Decoupling: Any client that can speak HTTP (e.g.,
curl, Python scripts, web apps, other LLM clients) can interact with the tools.Testing: Allows for direct API testing and debugging, independent of an LLM client.
AI agents
Once an AI agent is configured with access to rasdaman-mcp, it becomes capable of using several tools:
list coverages in the configured rasdaman
get the details of a particular coverage
execute processing/analytics queries based on a description in natural language
Examples
The following examples demonstrate the interaction with an AI agent using the rasdaman MCP server.
Listing Coverages

Describing a Coverage

Executing a Query

Query Result Visualization

Natural Language Query Suggestion

Development
Setup
Clone the git repository:
git clone https://github.com/rasdaman/rasdaman-mcp.git cd rasdaman-mcp/Create a virtual environment (if you don't have one):
uv venvActivate the virtual environment:
source .venv/bin/activateInstall from source:
uv pip install -e .
Core Components
Main Application (
main.py): This script initializes the FastMCP application. It handles command-line arguments for transport selection, rasdaman URL, username, and password. It then instantiates theRasdamanActionsclass and decorates its methods to expose them as tools.RasdamanActionsClass (rasdaman_actions.py): Encapsulates all interaction with the rasdaman WCS/WCPS endpoints. It is initialized with the server URL and credentials, and its methods contain the logic for listing coverages, describing them, and executing queries.WCPS crash course (
wcps_crash_course.py): A short summary of the syntax of WCPS, allowing LLMs to generate more accurate queries.
Defined Tools
The following methods are exposed as tools:
list_coverages(): Lists all available datacubes.describe_coverage(coverage_id): Retrieves metadata for a specific datacube.wcps_query_crash_course(): Returns a crash course on WCPS syntax with examples and best practices.execute_wcps_query(wcps_query): Executes a raw WCPS query and returns a result either directly as a string (scalars or small json), or as a filepath.
Documentation
To build the documentation:
# install dependencies
uv pip install '.[docs]'
sphinx-build docs docs/_buildYou can then open docs/_build/index.html in the browser.
Automated Tests
To run the tests:
# install dependencies
uv pip install '.[tests]'
pytestManual Testing
Interacting with the standalone HTTP server manually requires a specific 3-step process using curl.
The fastmcp protocol is stateful and requires a session to be explicitly initialized.
First, send an
initializerequest. This will return a200 OKresponse and, most importantly, a session ID in themcp-session-idresponse header (needed in the next steps).curl -i -X POST \ -H "Accept: text/event-stream, application/json" \ -H "Content-Type: application/json" \ -d '{ "jsonrpc": "2.0", "method": "initialize", "params": { "protocolVersion": "2024-11-05", "capabilities": {}, "clientInfo": { "name": "curl-client", "version": "1.0.0" } }, "id": 1 }' \ "http://127.0.0.1:8000/mcp"Next, send a notification to the server to confirm the session is ready. Use the session ID from Step 1 in the
mcp-session-idheader. This request will not produce a body in the response.SESSION_ID="<YOUR_SESSION_ID>" curl -X POST \ -H "Accept: text/event-stream, application/json" \ -H "Content-Type: application/json" \ -H "Mcp-Session-Id: $SESSION_ID" \ -d '{ "jsonrpc": "2.0", "method": "notifications/initialized" }' \ "http://127.0.0.1:8000/mcp"Finally, you can call a tool using the
tools/callmethod. Theparamsobject must contain thenameof the tool and anargumentsobject with the parameters for that tool. The server will respond with the result of the tool call in a JSON-RPC response.SESSION_ID="<YOUR_SESSION_ID>" # Example: Calling the 'list_coverages' tool curl -X POST \ -H "Accept: text/event-stream, application/json" \ -H "Content-Type: application/json" \ -H "Mcp-Session-Id: $SESSION_ID" \ -d '{ "jsonrpc": "2.0", "method": "tools/call", "params": { "name": "list_coverages", "arguments": {} }, "id": 2 }' \ "http://127.0.0.1:8000/mcp"
Available Tools
6 toolsdescribe_coverageA
Retrieves structural metadata for a specific datacube (coverage).
Inputs: coverage_id (required). Returns: result
execute_wcps_queryA
Executes a Web Coverage Processing Service (WCPS) query in rasdaman. Use this for spatio-temporal subsetting of datacubes, processing, aggregation, filtering.
Returns a structured dictionary indicating success, result_type, original query, the actual result value for scalar and small JSON or file path for large/binary results.
Important: Show the actual WCPS query and the result file path to the user.
Inputs: wcps_query (required). Returns: object
list_coveragesA
Lists all available datacubes (coverages) in rasdaman.
Inputs: no named parameters. Returns: result
wcps_doc_chaptersA
Lists the detailed WCPS documentation chapters, one per topic, each with a description of what it covers and when to fetch it. Use when the crash course is not detailed enough — pick the chapter matching your problem and fetch it with wcps_get_chapter(name). After a failed query, the 'common-errors' chapter maps error symptoms to fixes.
Inputs: no named parameters. Returns: result
wcps_get_chapterA
Returns one full WCPS documentation chapter (markdown) by name, as listed by wcps_doc_chapters(). Example: wcps_get_chapter("switch-case").
Inputs: name (required). Returns: result
wcps_query_crash_courseA
Returns a crash course on writing WCPS queries: learn the basic syntax, common operations, and best practices for WCPS queries. It's recommended to check this before executing queries.
Inputs: no named parameters. Returns: result
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 6 tools
Each tool maps to a distinct action: listing coverages, describing a coverage, learning WCPS basics, executing a query, listing doc chapters, and fetching a chapter. The documentation tools are explicitly tiered (crash course -> chapter index -> chapter body), so there is little risk of misselection.
The names are in readable snake_case, but they mix bare verb_noun names (list_coverages, describe_coverage, execute_wcps_query) with wcps_-prefixed resource-style names (wcps_query_crash_course, wcps_doc_chapters, wcps_get_chapter). The prefix and verb placement are inconsistent even though the overall style is still navigable.
Six tools is an appropriate size for a rasdaman querying server: coverage discovery, metadata inspection, query execution, and three levels of documentation. No tool feels redundant or missing at a high level.
The discover -> describe -> learn -> execute -> debug workflow is well covered. Minor gaps include a lack of write/ingest/update/delete operations for coverages and no explicit way to retrieve the contents of large result files, but these are likely outside the server's query-focused scope.
Maintenance
Related MCP Connectors
Astra DB tools for agents: schema, find, vector search, guarded writes, and interactive views.
Query, explore, and manage your databases through natural conversation.
List datasets, schemas, run APL queries, and use prompts for exploration, anomalies, and monitoring.
Discover, verify, and access NASA Earth science datasets via the Common Metadata Repository (CMR)
Related MCP Servers
- AlicenseAqualityDmaintenanceEnables natural language interaction with rasdaman databases, allowing users to list coverages, describe datacubes, and execute WCPS queries.5MIT
- AlicenseBqualityBmaintenanceEnables natural language interaction with a GeoServer instance for managing workspaces, datastores, feature types, layers, styles, and OGC services (WMS/WFS) via an LLM-powered agent.11MIT
- FlicenseNot gradedqualityDmaintenanceEnables interaction with Geo-INQUIRE OGC API Processes for geospatial operations, including planning, execution, and job management through natural language.-
- FlicenseNot gradedqualityDmaintenanceEnables LLMs to work with Copernicus Marine ocean data through semantic tools for dataset discovery, metadata retrieval, recommendation, validation, point time-series extraction, area statistics, subsetting, and computation without exposing raw API plumbing.-