How to Draft Email Responses from Slack with Make

Scenarios > Create new scenario > Slack

Create New Scenario

Start a new Make scenario to connect Slack message triggers with OpenAI response generation. This scenario will watch for new messages in a specific Slack channel.

1. 1Log into Make and click 'Create a new scenario'
2. 2Click the big + button in the center
3. 3Search for 'Slack' and select it from the app list
4. 4Choose 'Watch Messages' as your trigger

Slack module > Connection > Add

Connect Slack Account

Link your Slack workspace to Make so the scenario can access your channels and messages. You'll need admin permissions or pre-approval from your workspace admin.

1. 1Click 'Add' next to the Connection field
2. 2Select 'OAuth 2.0' connection type
3. 3Click 'Continue' and authorize Make to access Slack
4. 4Choose your workspace from the dropdown

Slack module > Configuration

Configure Channel Monitoring

Set which Slack channel to monitor for email content. Pick a dedicated channel like #email-drafts to avoid triggering on unrelated messages.

1. 1Select your workspace from the Team dropdown
2. 2Choose the target channel from the Channel dropdown
3. 3Set 'Limit' to 1 to process one message at a time
4. 4Leave 'Only messages from bots' unchecked

Add module > OpenAI > Create a Completion

Add OpenAI Module

Connect OpenAI's API to generate the email response. The GPT-4 model works best for professional email drafting compared to GPT-3.5.

1. 1Click the + button to the right of your Slack module
2. 2Search for 'OpenAI' and select it
3. 3Choose 'Create a Completion' action
4. 4Select 'GPT-4' from the Model dropdown

OpenAI module > Connection > Add

Connect OpenAI API

Add your OpenAI API key to authenticate requests. You'll need a paid OpenAI account since the free tier has strict rate limits.

1. 1Click 'Add' next to the Connection field
2. 2Paste your OpenAI API key in the API Key field
3. 3Click 'Save' to test the connection
4. 4Verify the green connection status appears

OpenAI module > Prompt field

Build the Prompt Template

Create a structured prompt that tells GPT-4 to draft a professional email response. The prompt should include the original email content from Slack.

1. 1Click in the Prompt field to open the mapping panel
2. 2Type: 'Draft a professional email response to the following:'
3. 3Press Enter twice and click 'Text' under Slack data
4. 4Select the message text field to insert the Slack message content

📬 New entry: {{1.name}}
Email: {{1.email}}
Details: {{1.description}}

OpenAI module > Parameters section

Configure Response Parameters

Set OpenAI parameters to generate appropriate email responses. Lower temperature gives more consistent professional tone.

1. 1Set Max Tokens to 300 for typical email length
2. 2Set Temperature to 0.3 for consistent professional responses
3. 3Leave Top P at default value of 1
4. 4Set Frequency Penalty to 0.2 to avoid repetition

Add module > Slack > Create a Message

Add Slack Response Module

Send the AI-generated response back to Slack so your team can review and use it. This creates a thread reply under the original message.

1. 1Click + after the OpenAI module
2. 2Search for 'Slack' and add another Slack module
3. 3Choose 'Create a Message' action
4. 4Use the same connection from step 2

Slack module > Message configuration

Configure Response Message

Set up the Slack message to post the generated response. Using a thread keeps the original email and response together.

1. 1Select the same channel as your trigger
2. 2Click in the Text field and select the OpenAI 'Text' output
3. 3Enable 'Reply in thread' toggle
4. 4Map the Thread TS field to the original message timestamp

📬 New entry: {{1.name}}
Email: {{1.email}}
Details: {{1.description}}

Scenario controls > Run once

Test the Scenario

Run a test to verify the complete workflow processes a message correctly. This checks all connections and data mapping.

1. 1Click 'Run once' at the bottom of the scenario
2. 2Go to your Slack channel and post a sample email
3. 3Return to Make and wait for the execution to complete
4. 4Check that a response appeared as a thread reply in Slack

OpenAI module > Right-click > Add error handler

Set Up Error Handling

Configure what happens when OpenAI is down or rate-limited. This prevents the scenario from breaking on temporary issues.

1. 1Right-click the OpenAI module and select 'Add error handler'
2. 2Choose 'Break' as the error handling directive
3. 3Add a filter to catch HTTP 429 (rate limit) errors specifically
4. 4Set a 2-minute delay before retry for rate limit errors

Scenario controls > Activate toggle

Activate the Scenario

Turn on the scenario to start monitoring your Slack channel continuously. Make will check for new messages every few minutes.

1. 1Click the toggle switch in the bottom left corner
2. 2Confirm activation in the popup dialog
3. 3Set the scheduling to run every 5 minutes
4. 4Save the scenario with a descriptive name like 'Email Response Drafter'

Drop this into a Make custom function.

Add this filter between Slack and OpenAI modules to only process messages that look like emails:
{{contains(1.text; "@")}} AND {{contains(1.text; "Subject:")}} OR {{contains(1.text; "From:")}}

Add this filter between Slack and OpenAI modules to only process messages that look like emails:

{{contains(1.text; "@")}} AND {{contains(1.text; "Subject:")}} OR {{contains(1.text; "From:")}}