Plugins

Steamship Plugins (opens in a new tab) perform specific tasks related to AI.

• How to use plugins
• How to develop plugins

Steamship supports the following types of plugins:

File Importers

Importers pull raw data from common external sources into a File.

Examples: A YouTube video importer imports video content given a URL, A Notion importer imports a document from a Notion space.

• Using File Importers
• Developing File Importers

Blockifiers

Blockifiers extract text and other content from raw data in a File to Blocks.

Examples: Whisper speech to text turns an audio file into a text transcript, a PDF extractor could pull the text chunks and images from a PDF document.

• Using Importers
• Developing Importers

Taggers

Taggers create Tags (annotations) on Files and Blocks.

Examples: A text classifier would attach a classification Tag to a Block, an image object recognizer would add Tags to a Block that identified known objects.

• Using Taggers
• Developing Taggers

Generators

Generators create new content from existing content.

Examples: GPT4 creates more text based on the existing text in a conversation, DALL-E creates an image based on a description.

• Using Generators
• Developing Taggers

Embedders

Embedders convert content into a vector representation. This is primarily used in combination with Steamship’s built in :ref:.

Examples: Use OpenAI to embed sentences into vectors for search; embed images into vectors for search

• Using Embedders
• Developing Embedders