Blockifying Data#

Blockifying content transforms raw bytes (audio, text, markdown, HTML) into Steamship’s Block and Tag Format format. This is how Steamship creates a space for many different AI models to interoperate over the same data.

Steamship’s plugin index maintains a list of Blockifiers you can use. Select the one appropriate for your data type and then apply it.

Here is how to blockify a Markdown file using the markdown-blockifier-default blockifier:

import { Steamship, MimeTypes } from "@steamship/client"
client = Steamship()

file = File.create(
   client=client,
   content="# This is a header \n\n And this is a paragraph!",
   mime_type=MimeTypes.MKD
)

blockifier = client.use_plugin("markdown-blockifier-default")
task = file.blockify(blockifier.handle)
task.wait()

Once Blockified, that file can be:

  • Refreshed, to pull down its blocks and tags locally,

  • Tagged, using tagger plugins, to add more information to it,

  • Queried using our query system