How to Auto-Translate Slack Messages with N8n and OpenAI

Workflows > New > Add Node > Slack Trigger

Create New N8n Workflow

Start a new workflow in N8n and add the Slack trigger node. This will listen for emoji reactions on messages in your workspace.

1. 1Click 'New workflow' in the N8n dashboard
2. 2Click the '+' button to add your first node
3. 3Search for 'Slack' and select 'Slack Trigger'
4. 4Choose 'Reaction Added' from the events dropdown

Node Settings > Credentials > Create New

Connect Slack Account

Link your Slack workspace to N8n using OAuth. You'll need admin permissions to install the N8n app in your workspace.

1. 1Click 'Create New' next to Credentials
2. 2Select 'Slack OAuth2 API' from the dropdown
3. 3Click 'Connect my account' and authorize in the popup
4. 4Select your workspace and click 'Allow'

Slack Node > Additional Fields > Filters

Configure Reaction Trigger

Set up the trigger to fire only on flag emoji reactions. This prevents the workflow from running on every reaction in your workspace.

1. 1In the Slack node, click 'Add Filter' under Additional Fields
2. 2Set Filter Key to 'reaction'
3. 3Set Filter Value to 'flag-*' to match any flag emoji
4. 4Enable 'Use Regex' checkbox for the wildcard pattern

Add Node > Function > Code Editor

Add Function Node for Language Detection

Create a function node to map flag emojis to language codes. This converts reactions like :flag-fr: into 'French' for OpenAI's API.

1. 1Click '+' to add a new node after Slack
2. 2Search for 'Function' and select it
3. 3Paste the flag-to-language mapping code in the code editor
4. 4Set the output to include both language name and original message text

Add Node > OpenAI > Chat

Connect OpenAI Translation Node

Add OpenAI node to handle the actual translation. Configure it to use GPT-3.5-turbo for cost efficiency since translation doesn't need GPT-4.

1. 1Add a new node and select 'OpenAI'
2. 2Choose 'Chat' as the resource type
3. 3Create new OpenAI credentials with your API key
4. 4Set Model to 'gpt-3.5-turbo' in the options

OpenAI Node > Messages

Configure Translation Prompt

Set up the chat prompt to translate messages accurately. Include context about preserving formatting and handling technical terms appropriately.

1. 1In the OpenAI node, set System Message to 'You are a translator. Translate the following message accurately, preserving any formatting or technical terms.'
2. 2Set User Message to reference the function output: 'Translate this to {{$node["Function"].json["language"]}}: {{$node["Function"].json["message"]}}'
3. 3Set Max Tokens to 500 to handle longer messages
4. 4Set Temperature to 0.1 for consistent translations

Add Node > Slack > Message > Post

Add Slack Reply Node

Configure a Slack node to post the translation as a thread reply. This keeps translations organized under the original message.

1. 1Add another Slack node after OpenAI
2. 2Set Resource to 'Message' and Operation to 'Post'
3. 3Use the same Slack credentials from step 2
4. 4Set Channel to reference the original channel: {{$node["Slack Trigger"].json["item"]["channel"]}}

Slack Message Node > Text and Thread Settings

Configure Thread Reply

Set the message to post as a threaded reply with the translation. Include the target language for clarity.

1. 1Set Text field to: 'Translation ({{$node["Function"].json["language"]}}): {{$node["OpenAI"].json["choices"][0]["message"]["content"]}}'
2. 2Set Thread TS to: {{$node["Slack Trigger"].json["item"]["ts"]}}
3. 3Enable 'As User' to false so it posts as the N8n bot
4. 4Test the workflow with Execute Workflow button

OpenAI Node > Settings > Error Handling

Add Error Handling

Configure error handling for API failures and unsupported languages. This prevents the workflow from breaking on edge cases.

1. 1Click Settings on the OpenAI node
2. 2Under 'On Error', select 'Continue With Default Value'
3. 3Set Default Value to 'Translation unavailable'
4. 4Enable 'Always Output Data' checkbox

Workflow > Activate > Test in Slack

Test with Real Reactions

Activate the workflow and test it with actual flag emoji reactions in Slack. Verify translations appear as threaded replies.

1. 1Click 'Save' then 'Activate' in the top right
2. 2Go to any Slack channel where the N8n app is installed
3. 3Post a test message and react with a flag emoji like :flag-es:
4. 4Check that the translation appears as a thread reply within 30 seconds

Drop this into an n8n Code node.

const flagToLanguage = {
  'flag-es': 'Spanish',
  'flag-fr': 'French', 

const flagToLanguage = {
  'flag-es': 'Spanish',
  'flag-fr': 'French', 
  'flag-de': 'German',
  'flag-it': 'Italian',
  'flag-pt': 'Portuguese',
  'flag-jp': 'Japanese',
  'flag-kr': 'Korean',
  'flag-cn': 'Chinese',
  'flag-ru': 'Russian'
};

const reaction = $input.item.json.reaction.replace(/:/g, '');
const language = flagToLanguage[reaction] || 'English';
const message = $input.item.json.item.message.text;

return { language, message, original_reaction: reaction };