Integration · Docs and data
AI agents for Google Document AI
Read documents with Google Cloud Document AI processors (OCR, forms, invoices, receipts, IDs, custom extractors).
- Actions
- 6
- Read
- 5
- Write
- 1
Actions
What your agents can do in Google Document AI
Read documents with Google Cloud Document AI processors (OCR, forms, invoices, receipts, IDs, custom extractors): process a file and get its text and extracted fields, start batch jobs over Cloud Storage and check them, and list the project's processors, signed in with a Google service account key.
Read 5
Look things up. Nothing changes.
Process a document
Reads one document (PDF, TIFF, GIF, JPEG, PNG, BMP or WebP) with a processor, from a Cloud Storage file (gs://bucket/path) or base64 content.
Check a batch job
Reads a batch job's state: done, metadata (state RUNNING, SUCCEEDED or FAILED, and per-document status with each output gs:// location) and error if it failed.
List processors
Lists the project's processors in this location, each with name (ending in its ID), type.
Get a processor
Reads one processor: type, displayName, state (ENABLED, DISABLED ...), defaultProcessorVersion, processEndpoint and createTime.
List processor types
Lists the kinds of processor this project may create in this location.
Write 1
Create, change or send.
Start a batch job
Starts reading many documents in Cloud Storage (every file under a gs:// folder prefix, or a list of files) with a processor, writing the results as JSON files under an output gs:// folder.
A person can approve first
Any write action can be set to wait for a person's approval before it runs, so an agent asks before it changes anything.
Connection
How it connects
Connect it once in the console, then give it to the agents and workflows that need it.
Enter the sign-in details Google Document AI gives you for its API.
What you enter
- Service account key (JSON)
Settings
- Project ID
- Location
Kept safe
- Credentials are stored encrypted with your organization's own keys.
- They are opened only by the process that makes the call, at the moment it runs, and never go into a prompt or the browser.
- Calls go only to the tool's own servers.
Ideas
Ideas to start with
Starting points made only from the actions above. Mix them with any other tool your agents can use.
- 01
Ask about your Google Document AI data
Ask a question in plain words. The agent can check a batch job, list processors or get a processor in Google Document AI, then answers with what it found.
- 02
Let an agent update Google Document AI
Have the agent start a batch job in Google Document AI as part of a larger task. Turn on approval and a person signs off on each change first.
- 03
Share what it finds in Slack
Give one agent Google Document AI and Slack: it can check a batch job in Google Document AI and post a summary to a Slack channel in the same run.
Works well with
More Docs and data tools
- Google Sheets3 actionsAdd rows to a Google Sheets spreadsheet, read a range of cells, and write over a range, signed in with the person's own Google account.
- Google Docs3 actionsCreate Google Docs documents, read their text, and add text to the end of them, signed in with the person's own Google account.
- Notion4 actionsAdd pages to Notion databases, append text to pages, query databases and search the pages shared with an internal integration.
- Confluence4 actionsCreate, read and search Confluence Cloud pages on your Atlassian site, signed in with an email address and an API token.
- Airtable4 actionsCreate, update, list and read records in Airtable bases that a personal access token can reach.
- Docusign3 actionsSend documents for signature from a Docusign template, check an envelope's status and list recent envelopes, as the person who connects.
FAQ
Google Document AI and MyDeepAgents
Can MyDeepAgents connect to Google Document AI?
Yes. The Google Document AI connector has 6 actions: 5 read and 1 makes changes. Your agents can call them as tools, and your workflows can run them as steps.
How do I connect Google Document AI?
Enter the sign-in details Google Document AI gives you for its API. You'll need: Service account key (JSON). The connection also asks for: Project ID and Location. Credentials are stored encrypted with your organization's own keys and are opened only at the moment of a call.
Can an agent change data in Google Document AI without asking?
Only if you let it. Its one action that changes data can be set to wait for a person's approval before it runs.
Can I use Google Document AI in workflows as well as agents?
Yes. Every Google Document AI action also works as a step in a workflow, and the same connection serves your agents and your workflows.
Early access
Put Google Document AI to work
Connect it in a few minutes, give an agent the actions it needs, and decide which changes wait for a person.