How to Create a Daily Productivity Report with n8n

What You Are Making

An n8n workflow that pulls yesterday's screen time data from ScreenTimerAI via MCP, feeds it to an AI model, and outputs a short daily productivity report.

You can schedule it to run every morning, send the result to Slack, email, or anywhere n8n can reach.

Why n8n

n8n is an open-source workflow automation platform that connects apps and services through a visual, node-based editor. Developers and teams use it to automate repetitive tasks without writing custom integrations from scratch. Each step in a workflow is a node you can drag, connect, and configure — no boilerplate code required.

Connecting screen time tracking to n8n turns a manual daily check-in into a fully automated pipeline. You set it up once and get a productivity report delivered to Slack, email, or any other destination every morning without lifting a finger. Because the workflow is visual, it is easy to customize the output, add filters, or route the report to multiple channels as your needs change.

Step 1: Install The MCP Community Node

The built-in n8n MCP Client node only supports SSE/HTTP transport. Since ScreenTimerAI uses stdio, you need the community node that adds stdio support.

In n8n, go to Settings, then Community Nodes, then Install. Enter:

n8n-nodes-mcp

You also need to set this environment variable for n8n (add it to your n8n startup config):

N8N_COMMUNITY_PACKAGES_ALLOW_TOOL_USAGE=true

Step 2: Create MCP Credentials

Go to Credentials, click New, and select "MCP Client (STDIO) API".

Fill in:

• Command: /Applications/ScreenTimerAI.app/Contents/MacOS/activity-mcp-server
• Arguments: (leave empty)
• Environment Variables: (leave empty)

Save the credential.

Step 3: Build The Workflow

Create a new workflow with these nodes:

Schedule Trigger

Set the trigger to run daily at your preferred time (e.g., 8:00 AM).

MCP Client Node

• Credential: Select the STDIO credential you created
• Operation: Execute Tool
• Tool Name: summarize_activity_range
• Parameters: Set the time range to yesterday

AI Agent Node (Optional)

Connect the MCP Client output to an AI Agent node to transform the raw data into a polished report. Use this system prompt:

You are a dark, witty productivity coach. Take the screen time data provided and write a ~150-word report. Focus more on failures than successes. Use short, punchy sentences. Swearing allowed sparingly for impact.

Output Node

Send the result wherever you want: Slack message, email, webhook, or just save it to a file.

Other Prompts To Try

The daily report workflow is a starting point. Here are four variations you can build with the same MCP connection:

• App usage breakdown workflow: Use the get_activity_logs tool to fetch raw activity data and pipe it through a Code node that groups entries by app name. The output is a ranked list of your most-used apps, ready to send to Slack or save to a spreadsheet.
• Weekly review automation: Replace the daily Schedule Trigger with a weekly one and call summarize_activity_range with a seven-day window. Format the result as a digest email that lands in your inbox every Monday morning.
• Focus score alert: Use show_focus_score_timeline on a recurring schedule and add an IF node that checks whether the focus score drops below a threshold you define. Only trigger a notification when it does — no spam on good days.
• Standup prep: Schedule the workflow to run every morning before your standup meeting. Call summarize_activity_range for yesterday and use a Code node or AI Agent to format the summary as bullet points, then post them to a Slack channel so your team sees what you shipped.

Troubleshooting

• Community node not appearing? Make sure N8N_COMMUNITY_PACKAGES_ALLOW_TOOL_USAGE=true is set in your n8n environment variables. Restart n8n after adding or changing this value.
• STDIO credential failing? Verify the binary path is correct and accessible from where n8n runs. This is especially important if n8n is running in Docker — the ScreenTimerAI binary must be mounted into the container at the path you specify in the credential.
• AI Agent not calling tools? Attach the MCP Client node as a sub-node to the AI Agent node, not as a standalone node in the workflow. The AI Agent only discovers tools from its connected sub-nodes.

What Happens Next

Once the workflow is active, n8n runs it on schedule. Every morning you get a fresh productivity report delivered wherever you want.

No manual steps. No prompting. Fully automated.