ironclaw/docs/extensions/google/docs.md

title, description

title	description
Docs	Let your agent create and edit Google Documents

The Google Docs extension allows your agent to interact with Google Docs — creating documents, inspecting paragraphs and tables, applying anchored edits, creating populated tables, verifying results, and using low-level formatting or batch updates when needed. Semantic operations validate inputs and read provider state back, reducing multi-call index discovery loops.

---

Setup

If you haven't set up Google OAuth yet, complete the Google OAuth Setup first.

In your Google Cloud project, navigate to APIs & Services → Library, search for Google Docs API, and click Enable.

ironclaw extension install google-docs

ironclaw extension activate google-docs

Activating an extension that needs credentials starts its setup flow. Complete it from Extensions in the web interface — for Google services that means the OAuth consent screen. Follow the auth setup first so your agent can capture the callback. Tokens are stored encrypted and refreshed automatically.

If you already authenticated one Google service, you still need to authenticate each additional Google extension separately.

---

Available Actions

• create_document: Create a new Google Doc with an optional title
• get_document: Retrieve document metadata (title, revision, named ranges)
• read_content: Extract the plain-text or structured content of a document
• insert_text: Insert text at a specific index in the document body
• delete_content: Delete a range of content by start and end index
• replace_text: Find and replace text throughout the document
• format_text: Apply character formatting (bold, italic, font size, color) to a text range
• format_paragraph: Apply paragraph styling (heading level, alignment, spacing, indentation) to a range
• insert_table: Insert a table with a specified number of rows and columns
• create_list: Convert a range of paragraphs into a bulleted or numbered list
• batch_update: Send multiple document update requests in a single API call
• inspect_document: Inspect paragraphs and tables with stable document indexes
• apply_text_edits: Apply validated, tab-scoped text replacements and verify the result
• create_table_with_data: Insert, populate, style, and verify a table
• verify_document: Check expected text and specific table contents against provider state

---

Example Usage

Once configured, you can ask your agent things like:

• "Create a new document titled 'Q2 Marketing Plan'"
• "Read the content of document ID 1BxiMVs0XRA5nFMdKvBdBZjgmUUqptlbs74OgVE2upms"
• "Insert a summary paragraph at the top of my report"
• "Replace all occurrences of 'TBD' with 'Pending Review' in this doc"
• "Format the title as Heading 1 and make it bold"
• "Add a 3-column table for the budget breakdown"

---

Working with Document IDs

Google Doc IDs appear in the document URL:

https://docs.google.com/document/d/<DOCUMENT_ID>/edit

You can tell your agent to "use the document at this URL" and paste the full URL — the agent will extract the document ID automatically.