Does AIND Data offer any AI-powered features?

Yes, AIND Data includes a `get_summary` tool that leverages AI to generate concise summaries for specific data assets, helping users quickly understand complex datasets.

How does AIND Data help navigate the V2 metadata schema?

It provides dedicated tools such as `get_top_level_nodes` to explore schema fields and `get_modality_types` to list available data modality names, making schema navigation intuitive for users.

Is AIND Data designed for integration with developer environments?

Absolutely. AIND Data is built as an MCP (Model Context Protocol) server, making it compatible and configurable for use in various developer environments and AI clients like VSCode, Cursor, and the Claude Desktop App.

What kind of data can I query and retrieve using AIND Data?

You can query MongoDB collections with filters and projections, execute complex MongoDB aggregation pipelines, explore V2 metadata schema details, and access NWB (Neurodata Without Borders) files directly from Code Ocean or S3 storage.

What is AIND Data and its primary purpose?

AIND Data is an MCP server providing a comprehensive set of tools to access metadata and data assets from the Allen Institute for Neural Dynamics (AIND), specifically engineered to work with the V2 aind-data-schema format.

AIND Data: Access & Query Neuroscience Data (V2 Schema)

The AIND Data tool is a Model Context Protocol (MCP) server designed to facilitate interaction with the Allen Institute for Neural Dynamics' (AIND) data and metadata, specifically adhering to the V2 aind-data-schema. It offers a comprehensive suite of functionalities for data retrieval and querying, including filtering MongoDB collections, executing complex aggregation pipelines, and counting records. Additionally, it provides robust schema navigation tools, allowing users to explore top-level fields, get query-writing guidance, and list data modality types. The server also includes schema examples for various document types and specialized tools for accessing NWB files from Code Ocean and S3. It integrates seamlessly with various MCP clients like VSCode extensions and Cursor IDE, serving as a vital bridge for neuroscientific data access and exploration.

Características Principales

01Query MongoDB collections with filters and projections using `get_records`

02Generate AI-powered summaries for data assets using `get_summary`

03Access NWB files from Code Ocean (`identify_nwb_contents_in_code_ocean`) or S3 (`identify_nwb_contents_with_s3_link`)

040 GitHub stars

05Execute complex MongoDB aggregation pipelines via `aggregation_retrieval`

06Navigate V2 metadata schema with tools like `get_top_level_nodes` and `get_modality_types`

Casos de Uso

01Programmatically exploring AIND data schemas and retrieving specific document examples for development and analysis

02Accessing and querying AIND V2 metadata and data assets in AI-powered development environments

03Integrating neuroscientific data access into MCP-compatible IDEs like VSCode (Cline, Github Copilot) and Cursor