Server facts
tigris-mcp-server
Project
- Author
- tigrisdata
- Repository
- github.com/tigrisdata/tigris-mcp-server
- Website
- www.tigrisdata.com
- License
- MIT
- Category
- Cloud & DevOps
Tech
- Runtime / language
- node
- Transports
- stdio
- Remote available
- No
- Authentication
- Not specified
Activity
- GitHub stars
- 4
- Last commit
- Data as of
Overview
Tigris storage management can be part of an AI editor workflow through this MCP integration. Agents can inspect the account’s buckets, create buckets for new work, or delete them. Within a bucket, the tools support listing, retrieving, creating, and deleting objects. Content can be supplied directly or uploaded from a local filesystem path. Sharing workflows have dedicated operations for generating an object’s signed URL and for uploading a file while returning a public URL. These operations let developers work with stored files and bucket organization from their AI client.
The server communicates over stdio and has a Node.js runtime, with documented launch options using NPX or Docker. Setup guidance covers VS Code, Claude Desktop, and Cursor. Authentication can use access keys or an existing AWS profile when the AWS CLI and credentials are configured. The Docker instructions describe restricting filesystem operations to a mounted directory. The project also points users to a hosted Tigris MCP service with OAuth support.
tigris-mcp-server setup notes
- tigris-mcp-server records stdio only; confirm a network transport upstream before planning remote access.