FlowRunner
Features Pricing
Theme
Start Free

Google Docs

Documents

Create, read, and edit Google Docs documents through the Google Docs API. Agents fill placeholder tokens with workflow data to produce contracts, invoices, and reports, and apply text and formatting edits in batched updates.

Verified 7 actions OAuth available
Google Docs website Platform Documentation Capability data verified 2026-10-21
A record reaches the stage that needs a document (deal closed, hire approved, report due)
Agent pulls the field values the template expects from the source system
Google Drive Copy File duplicates the template first
Replace Placeholders fills every {{placeholder}} in the copy
Agent checks the returned occurrence count so no token was left unfilled
Google Drive Export File saves the finished doc as a PDF
Agent posts the stored PDF link to the owning channel
Client-facing contracts pause for an owner to approve before the PDF is sent outside the company

What This Integration Enables

Agents can create, read, and edit Google Docs as steps inside a larger workflow. The most common pattern is document generation: copy a template with Google Drive, fill its tokens with Replace Placeholders, export the result to PDF through Google Drive, and pass it downstream. Agents can also read a document's full plain text for summarization or AI processing, append content to a running log, and reach any advanced Docs operation through the raw Batch Update passthrough when a dedicated action does not cover it.

The point is not that FlowRunner can write to a document. The point is that generating the document stops being a person's manual chore and becomes a governed step, with a human pulled in only when the finished document is about to leave the building.

Without FlowRunner

Manual template copying Someone opens the template, copies fields in by hand, and hopes nothing was missed
Silent unfilled tokens A stray {{placeholder}} ships in the final document with no check that every field was replaced
Scattered exports Each finished document is exported and saved wherever the person happened to be working

With FlowRunner

Generated from record data The agent fills every placeholder straight from the source system in one step
Verified token fill The occurrence count confirms each placeholder was replaced before the document moves on
Consistent PDF handoff Every document lands in FlowRunner file storage as a PDF ready for the next step

Flows you can build with Google Docs

    • Google Docs
    • DocuSign

    Contract generated from deal data

    A deal moves to the stage that needs a signed agreement. The agent reads the deal record from HubSpot, copies the agreement template with Google Drive Copy File, then calls Replace Placeholders on the copy with the deal's company name, term, and pricing mapped to the template's {{placeholders}}. It checks the returned occurrence count to confirm every token was filled, then runs Google Drive Export File to produce a PDF in FlowRunner file storage. The finished contract is ready to route for signature through DocuSign. The work shifts from a rep hand-editing a Word file to an agent producing a consistent, verified document every time.

    • Google Docs
    • Slack

    Meeting notes summarized and shared

    A recurring meeting document needs a digest. The agent calls Get Document, which returns a convenience text field with the full body as plain text. It hands that text to an AI summarization step, then posts the summary to the team's Slack channel. Nobody copies the doc into a chat window and trims it down by hand. The running document stays the source of record and the summary reaches the people who need it.

    • Google Docs
    • Google Forms
    • Gmail

    Personalized welcome letters with a human gate

    A new customer submits an onboarding form captured through Google Forms. The agent generates a personalized welcome letter by copying the template and running Replace Placeholders on the copy, then exports it to PDF through Google Drive. Because the letter carries the company's name and commitments to an outside party, the agent does not send it automatically. It pauses and asks the account owner to approve the wording. Once approved, the agent emails the PDF through Gmail. The routine drafting is automated; the judgment about what goes out under the company's name stays with a person.

Human-in-Loop Highlight

The consequential moment with Google Docs is not creating a document, it is sending a generated document to someone outside the company. When an agent produces a client contract or an offer letter from a template, it pauses before that document leaves the building and asks the owner through Slack: "The [Company] agreement is generated and exported to PDF. Term is 12 months, total is the value pulled from the deal record. Approve to send for signature, or flag a field to fix?" The owner approves or corrects. A generated document that is wrong in a placeholder is a document that goes out wrong to a customer, so the agent treats the external send as the point that needs human judgment, not the drafting that led up to it.

Agent processes routinely
Detects exception requiring judgment
Clear match Continues automatically
Ambiguous Routes to human via Slack
Human decides
Agent resumes with decision

Agent Capabilities

7 actions

Documents

2
  • Create Document Creates a new blank Google Docs document with a given title in the connected user's My Drive root, optionally inserting initial body text. Used when a workflow needs a fresh document to build up, such as a running log or a draft the agent will fill in later steps.
  • Get Document Retrieves a document by ID and returns the raw Docs resource plus a convenience `text` field with the full body as concatenated plain text. Used when a workflow needs to read a document for summarization, analysis, or AI processing, or needs the body structure for index-based editing.

Text Editing

4
  • Append Text Appends text to the very end of a document body, using newline characters to create new paragraphs. Used to keep a running document current, such as adding each workflow run summary to a shared log.
  • Insert Text Inserts text at a specific character index in the document body. Used when content must land at a known position rather than at the end, using the index bounds returned by Get Document.
  • Replace All Text Replaces every occurrence of a text string with replacement text, with case-sensitive or case-insensitive matching. Used to update recurring values across a document or, with an empty replacement, to strip content out.
  • Batch Update Executes a raw Google Docs batchUpdate request, the escape hatch for operations not covered by dedicated actions such as styling, tables, images, named ranges, and headers. Used when a workflow needs an advanced Docs operation applied atomically.

Templates

1
  • Replace Placeholders Fills `{{placeholder}}` tokens in a document from a key and value map in one atomic batch update, with case-sensitive matching, and returns an occurrence count per placeholder so the agent can prove every token was filled. It edits the document it is given, so generation flows copy the template first with Google Drive Copy File and fill the copy. The primary action for generating contracts, invoices, offer letters, and personalized documents from workflow data.

Frequently asked questions

What can FlowRunner do with Google Docs?

FlowRunner agents can run Create Document, Get Document, and Append Text in Google Docs, plus 4 more actions.

Does connecting Google Docs to FlowRunner require OAuth?

Yes. Google Docs connects to FlowRunner with OAuth 2.0, so agents authenticate without handling raw credentials.

Can Google Docs trigger a FlowRunner workflow automatically?

Google Docs doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger.

Start building with Google Docs

Free plan, no card required. Connect in minutes.