Google Docs Composio MCP integration for AI agents

<h2>Google Docs Composio MCP & Chase Agents</h2><h3>Simplify Document Automation: Chase Agents & Google Docs Composio MCP Integration</h3><p>Chase Agents streamlines your document workflows further by integrating with the Google Docs Composio MCP. This combination offers pre-built actions and an easy-to-use interface for creating, reading, and updating Google Docs within automated processes. Build agents and applications that manage document tasks effortlessly, without writing code, using Chase Agents's intuitive platform.</p>

Connect Google Docs Composio MCP to Chase Agents as an MCP server, then use it in plain-English automations — on the web, in Slack, or inside ChatGPT and Claude via MCP. Chase Agents builds the workflow and deterministic code runs it.

How to connect Google Docs Composio MCP

Add Google Docs Composio MCP from the Chase Agents connections marketplace as MCP server, provide the required credentials, and reference it in your automation steps. See plans or read the connection guide.

About Google Docs Composio MCP

When getting documents by document ID, you pass in the document's id as the field "id", not as "document_id".

Document Copy Operations

To copy an existing Google Document using GOOGLEDOCS_COPY_DOCUMENT:

Required Parameters:

  • document_id (string): The ID of the document to copy (from URL: docs.google.com/document/d/{ID}/edit)
  • title (string): The title for the new copied document

Example:

{
  "tool_name": "GOOGLEDOCS_COPY_DOCUMENT",
  "arguments": {
    "document_id": "1-Ba-0Ro0TubT10oxfyniCxw3Z618mdkNYrcvHjzFHig",
    "title": "New Document Name"
  }
}

Returns: The new document with its ID, display URL, and confirmed title name. Location: Document is placed in the user's root Google Drive folder by default.

Efficient Text Extraction from Google Docs

When using GOOGLEDOCS_GET_DOCUMENT_BY_ID, the response provided to sandbox code is often structured as a list, where the actual data resides in element 0's 'text' field as a stringified JSON.

Mistakes to avoid:

  1. Blindly assuming the tool output is a direct dictionary you can access via ['data'].
  2. Incorrectly assuming file paths are directly accessible in sandbox code (FileNotFoundError).
  3. Guessing indexing without inspecting structure.

Correct Workflow:

  1. Call GOOGLEDOCS_GET_DOCUMENT_BY_ID.
  2. Access the output list: data_string = dataPrev[0]['text'].
  3. Parse the stringified JSON: doc_json = json.loads(data_string).
  4. Navigate to content: doc_data = doc_json['data'].
  5. Iterate through content: doc_data['body']['content'] → paragraph → elements → textRun.
  6. Concatenate strings from textRun['content'].

This approach bypasses path-finding issues and handles the data structure directly.

Official Google Docs Composio MCP documentation