Slack Research Assistant n8n Workflow: OpenAI + Apify MCP

Get fast, source-linked answers in Slack when someone asks a question

This Slack Research Assistant n8n workflow listens for Slack @mentions, runs an OpenAI-powered research agent with Apify web research tools (via MCP), and posts a concise, URL-backed answer directly back in the same Slack thread.

What this workflow does

• Triggers on mentions: When the Slack app is mentioned in a channel your bot has joined, the workflow starts (Slack app_mention events).
• Extracts the question: It cleans the mention text to isolate the research question the user asked.
• Plans and keeps context: An OpenAI chat model with thread-based memory helps the agent plan research and handle follow-up questions in context.
• Collects live web sources: It calls Apify MCP tools to gather information, including actors such as the RAG Web Browser and Google Search Results Scraper (plus other Apify Store Actors as configured).
• Synthesizes and formats: It produces a Slack-formatted response with supporting URLs.
• Replies in-thread: The answer is posted back as a reply in the original message thread.

Use cases

• Answering “What are the latest best practices for X?” with fresh web sources.
• Supporting SaaS operators with research for product decisions, competitor comparisons, or market questions.
• Helping teams move faster during incident reviews by summarizing verified information found online.

Technical details

• Slack Trigger + reply: Requires Slack scopes app_mentions:read and chat:write.
• OpenAI agent: Uses n8n nodes n8nn8n-nodes-langchainagent and n8nn8n-nodes-langchainlm chat open ai with thread-based memory.
• Apify via MCP: Uses n8nn8n-nodes-langchainmcp client tool with an Apify API token (HTTP Bearer credential) attached to the MCP tools endpoint.
• Setup essentials: Provide Slack Bot Token, OpenAI API key, and Apify token; register Event Subscriptions for app_mention; paste the system message context for accurate “us/our” answers.