Lium connects to the systems your work already lives in: object storage, warehouses and databases, documentation and issue trackers, and public scientific catalogs.
The named vendors below are the ones we have wired for you. Everything else you bring yourself — over MCP, a generic HTTP API, or your own bucket and warehouse credentials.
31 connections
Named services we have already wired. Authenticate and go.
Read Jira issues and Confluence pages through Atlassian's Rovo MCP server.
Search AWS documentation, regional service availability, and architectural guidance.
Search Cloudflare product documentation and migration guides.

Fetch up-to-date, version-specific documentation and code examples for libraries and frameworks.
Ask questions about and read AI-generated documentation for public GitHub repositories.
Look up protein families, domains, and functional site annotations from InterPro.
Search and traverse biomedical ontologies and their terms, hosted by EMBL-EBI.
Read Figma files, components, and design metadata.
Search and manage repositories, issues, and pull requests.
Search models, datasets, and Spaces on the Hugging Face Hub.
Read issues, projects, and cycles from Linear.
Search official Microsoft and Azure documentation and training content.

Search NASA Earth observation datasets and granules through the Common Metadata Repository.
Read and search Notion pages, databases, and comments.

Explore gene-disease associations, drug targets, and supporting evidence.

Query product analytics, feature flags, session recordings, and experiments.

Search academic papers, authors, and citation graphs across the scientific literature.
Read payments, customers, invoices, and subscriptions, and search Stripe documentation.
Inspect Supabase project schemas, tables, and logs. Development and testing projects only -- Supabase advises against pointing this at production data.

Query World Bank development indicators and country statistics.
Point Lium at your own infrastructure — your bucket, your warehouse, your endpoint, your MCP server.
Buckets in AWS S3, including public datasets and private AWS accounts.
Containers in Azure Storage accounts using public, SAS, or account-key access.
Catalogs, schemas, and tables served by a Databricks SQL warehouse.
Connect a Dropbox account and browse files to ingest.
Datasets and tables in Google BigQuery.
Buckets in Google Cloud Storage with public or service-account access.
REST APIs and SaaS platforms reached over HTTP, with or without authentication.
Connect a remote MCP server and expose its tools to your agents.
Tables in a MySQL database, using a stored username and password or short-lived AWS RDS IAM tokens.
Tables in a PostgreSQL database, using a stored username and password or short-lived AWS RDS IAM tokens.
Databases, schemas, and tables in a Snowflake cloud data warehouse.
This list is a convenience layer, not a boundary. Anything that speaks MCP, exposes an HTTP API, or lives in your own object storage or warehouse connects without waiting for us to add it — which is most of what teams actually need to reach.
Point Lium at an MCP server, a generic HTTP endpoint, or your own bucket, database or warehouse. Your credentials stay yours.

Join the leaders accelerating insights with real-world data.