- Home
- Skills
- Workflow Automation
- Google Workspace Automation
Works with the AI Agents you already use
Google Workspace Automation
Managing Google Workspace manually is a time sink that disrupts development flow.
Free
Google Workspace Automation
Example session with this skill installed
Search for unread emails from the last 7 days about "Project X", summarize them, and draft a response to the most recent one.
- Read your context and instructions
- Compiled the google workspace automation
I have searched your Gmail for unread threads related to "Project X" from the past week. I found 3 relevant threads and prepared a summary and a draft response for the latest message.
Gmail Automation Summary
| Thread Subject | Sender | Date | Key Request / Action |
|---|---|---|---|
| Re: Project X Timeline | sarah.engineer@company.com | 2023-10-24 | Requested approval on the Phase 2 architecture diagram. |
| Project X: Missing Assets | design-team@company.com | 2023-10-23 | Notified that the SVG icons for the landing page are missing. |
| Budget Update - Project X | finance-dept@company.com | 2023-10-21 | Sent the revised Q4 spreadsheet for review. |
Proposed Draft Response
To: sarah.engineer@company.com
Subject: Re: Project X Timeline
Hi Sarah,
I've reviewed the Phase 2 architecture diagram you sent over. The transition logic for the authentication flow looks solid. I have one minor question regarding the token refresh interval, but otherwise, I'm ready to approve. Let's discuss briefly during the standup, or I can send the final approval via the portal.
Best,
[User]
Next steps
- Provide a "yes" to send the draft to Sarah.
- Let me know if you want me to search for the missing SVG icons mentioned by the design team.
- I can read the Q4 budget spreadsheet from the finance email if you need a summary.
Connects securely to your tools. The creator never sees your data.
About this skill
Managing Google Workspace manually is a time sink that disrupts development flow. This skill automates Gmail, Calendar, Drive, Docs, and Sheets by interfacing with connected tools, CLIs, or Python APIs. It handles the overhead of searching threads, scheduling meetings across time zones, and processing spreadsheet data so you can stay in your IDE.
What it does
- Gmail orchestration searches with advanced operators, summarizes threads, and manages drafts.
- Calendar management finds free/busy slots for multiple attendees and suggests meeting options.
- Drive organization locates files by metadata and proposes structured folder migrations.
- Spreadsheet processing reads A1 ranges, appends rows, and generates reports with Python aggregation.
- Document generation clones templates and replaces placeholders while preserving formatting.
- Apps Script deployment builds and installs time-driven automation scripts for recurring tasks.
How it works
- Access discovery checks for connected MCP tools or local CLIs like
gwsorgog. - Environment setup installs necessary Python libraries and manages OAuth tokens with narrow scopes.
- Execution flow performs read operations to gather context from emails, docs, or sheets.
- Safety check presents a draft, table, or summary of proposed changes for user approval.
- Finalization executes the modification and returns specific IDs and links for the created resources.
Frameworks & tools
Uses google-api-python-client and google-auth-oauthlib for API access. Supports gws and gog CLIs.
Why this beats prompting it yourself
Generic prompts often struggle with specific Gmail search operators or complex A1 range logic in Sheets. This skill enforces a "Draft first, confirm later" safety protocol and handles the technical boilerplate of OAuth scopes and Python data aggregation.
Use cases
- Generate a daily email digest of unread threads excluding promotions.
- Find common free time for three stakeholders and draft the calendar invites.
- Batch process a CSV into an existing Google Sheet and update summary charts.
- Build a folder structure in Drive based on a project list and move files.
Known limitations
Does not support silent OAuth scope expansion. Requires manual token authorization via a desktop client for Python API access.
Requires local Python environment for API scripts. OAuth requires manual browser-based authentication. Cannot bypass "Draft first" confirmation for destructive or outbound actions.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
This skill has been used 7 times.
Trust & safety
Security scan
Security scan passed 7 days ago
- Free with an Agensi account
Requires
Permissions
No special permissions declared or detected