Fireflies to Slack, meeting recaps everyone reads
Meeting notes are where good intentions go to disappear. Someone says “I’ll send a recap,” then Friday hits, the transcript is buried, and nobody remembers who owned what.
Project managers feel it first. Then the marketing lead trying to keep launches moving, and the ops person who’s tired of chasing updates in threads. This Fireflies Slack recaps automation turns finished transcripts into a clean Slack recap plus ClickUp tasks, usually within minutes of the call ending.
Below, you’ll see exactly what the workflow does, what it replaces, and how it keeps decisions and follow-ups visible without extra admin work.
How This Automation Works
See how this solves the problem:
n8n Workflow Template: Fireflies to Slack, meeting recaps everyone reads
flowchart LR
subgraph sg0["Summary Generator Flow"]
direction LR
n0@{ icon: "mdi:cog", form: "rounded", label: "Get a transcript", pos: "b", h: 48 }
n1@{ icon: "mdi:swap-vertical", form: "rounded", label: "Split Out", pos: "b", h: 48 }
n2@{ icon: "mdi:cog", form: "rounded", label: "Aggregate", pos: "b", h: 48 }
n3@{ icon: "mdi:brain", form: "rounded", label: "Google Gemini Chat Model", pos: "b", h: 48 }
n4["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/clickup.svg' width='40' height='40' /></div><br/>Create a task"]
n5["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/slack.svg' width='40' height='40' /></div><br/>Send a message"]
n6["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/webhook.dark.svg' width='40' height='40' /></div><br/>Firefly Webhook"]
n7@{ icon: "mdi:robot", form: "rounded", label: "Summary Generator", pos: "b", h: 48 }
n8["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/code.svg' width='40' height='40' /></div><br/>Cleans AI response"]
n9["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/code.svg' width='40' height='40' /></div><br/>Extracts action items"]
n2 --> n7
n1 --> n2
n6 --> n0
n0 --> n1
n7 --> n8
n8 --> n5
n8 --> n9
n9 --> n4
n3 -.-> n7
end
%% Styling
classDef trigger fill:#e8f5e9,stroke:#388e3c,stroke-width:2px
classDef ai fill:#e3f2fd,stroke:#1976d2,stroke-width:2px
classDef aiModel fill:#e8eaf6,stroke:#3f51b5,stroke-width:2px
classDef decision fill:#fff8e1,stroke:#f9a825,stroke-width:2px
classDef database fill:#fce4ec,stroke:#c2185b,stroke-width:2px
classDef api fill:#fff3e0,stroke:#e65100,stroke-width:2px
classDef code fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px
classDef disabled stroke-dasharray: 5 5,opacity: 0.5
class n7 ai
class n3 aiModel
class n6 api
class n8,n9 code
classDef customIcon fill:none,stroke:none
class n4,n5,n6,n8,n9 customIcon
The Challenge: Meeting follow-ups fall through
Fireflies gives you a transcript, but a transcript isn’t a recap. It’s long, messy, and nobody wants to skim 40 minutes of talk to find the two decisions that mattered. So someone volunteers to summarize, copy key bits into Slack, and manually create tasks in ClickUp. That sounds reasonable until you do it after three calls in one day, miss a detail, or assign the wrong owner. The real cost isn’t just time. It’s the confusion that shows up later when people disagree on what was decided.
It adds up fast. Here’s where it breaks down in real teams.
- The transcript gets posted, but no one reads it because it’s too long to scan.
- Action items live in someone’s private notes, which means follow-ups turn into awkward “hey, just checking…” messages.
- Creating ClickUp tasks manually takes long enough that people skip it when they’re busy.
- Recaps vary wildly by person, so the team stops trusting that Slack message as the source of truth.
The Fix: Fireflies transcript → Slack recap + ClickUp tasks
This workflow starts the moment Fireflies finishes a transcription and sends a notification. n8n receives that webhook, pulls the transcript using the meeting ID, and turns the raw content into something AI can reliably summarize by splitting it into sentences and re-combining it into a clean text block. Then a Gemini-powered AI agent generates two outputs in one pass: a short, readable meeting summary and a structured list of action items (with titles and descriptions). Finally, the workflow cleans the AI response into predictable JSON, posts the recap in Slack, and creates each action item as a ClickUp task so nothing gets lost. You stop “writing notes” and start reviewing them.
The flow begins with a Fireflies webhook and transcript retrieval. AI turns that transcript into decisions plus next steps. Slack gets the recap immediately, and ClickUp receives tasks people can actually work from.
What Changes: Before vs. After
| What This Eliminates | Impact You’ll See |
|---|---|
|
|
Real-World Impact
Say your team runs 10 meetings a week and you want a recap plus tasks every time. Manually, it’s usually about 20 minutes to skim the transcript, 10 minutes to write a decent Slack recap, and another 10 minutes to create tasks in ClickUp, so roughly 40 minutes per meeting. That’s about 6 to 7 hours weekly. With this workflow, the “work” is basically waiting for Fireflies to finish, then reviewing what gets posted in Slack; most teams spend maybe 5 minutes per meeting doing quick edits, which is under an hour a week.
Requirements
- n8n instance (try n8n Cloud free)
- Self-hosting option if you prefer (Hostinger works well)
- Fireflies.ai to access meeting transcripts by ID.
- Slack to publish recaps where your team already works.
- ClickUp to create tasks automatically from action items.
- Google Gemini API (PaLM credentials) (get it from Google AI Studio / Google Cloud console).
Skill level: Intermediate. You’ll connect OAuth accounts and paste API credentials, plus do light field mapping for ClickUp.
Need help implementing this? Talk to an automation expert (free 15-minute consultation).
The Workflow Flow
Fireflies finishes a transcript and calls your webhook. Once the meeting transcription is done, Fireflies sends n8n the meeting ID, so the workflow knows exactly what to fetch.
The transcript is retrieved and cleaned up for summarization. n8n pulls the transcript, splits it into sentences, then recombines it into a single text block. That sounds small, but it helps the AI produce more consistent summaries.
Gemini generates a summary and action items. An AI Agent runs against Google Gemini to produce a short recap plus a structured list of tasks. The workflow then cleans the response into JSON so downstream steps don’t break when formatting changes.
Slack gets the recap, ClickUp gets the work. The formatted summary is posted to Slack, and each extracted action item becomes a ClickUp task with a title and description.
You can easily modify the Slack message format to match your channel norms, or change how ClickUp tasks are titled based on meeting type. See the full implementation guide below for customization options.
Step-by-Step Implementation Guide
Step 1: Configure the Webhook Trigger
Set up the inbound webhook that receives Fireflies transcription completion events.
- Add and open Incoming Fireflies Hook.
- Set HTTP Method to
POST. - Set Path to
fireflies-n8n-core47-agent. - Copy the generated webhook URL and paste it into your Fireflies webhook configuration.
Step 2: Connect Fireflies and Retrieve Transcripts
Use the meeting ID from the webhook payload to fetch the transcript details.
- Add Retrieve Transcript and connect it to Incoming Fireflies Hook.
- Set Transcript ID to
={{ $json.body.meetingId }}. - Credential Required: Connect your firefliesApi credentials.
Step 3: Prepare the Transcript for Summarization
Split sentences and aggregate them into a single raw transcript string for the AI prompt.
- Add Separate Sentences and connect it to Retrieve Transcript.
- Set Field To Split Out to
data.sentences. - Add Combine Text and connect it to Separate Sentences.
- In Combine Text, aggregate the raw_text field (default aggregation is used).
Step 4: Set Up AI Summarization
Generate a concise meeting summary and action items using Gemini as the language model.
- Add Generate Summary and connect it to Combine Text.
- Set Prompt Type to
define. - Set Text to the full prompt in the node, including the expression
{{ $json.raw_text }}. - Add Gemini Chat Engine as the language model for Generate Summary and set Model Name to
models/gemini-2.0-flash-thinking-exp-1219. - Credential Required: Connect your googlePalmApi credentials in Gemini Chat Engine (credentials are added to the parent AI model node, not Generate Summary).
Step 5: Clean AI Output and Route in Parallel
Parse the AI JSON output, then send the summary to Slack and extract action tasks simultaneously.
- Add Clean AI Output and connect it to Generate Summary.
- Set JavaScript Code to the provided parser that removes code fences and parses JSON.
- Clean AI Output outputs to both Post Slack Summary and Extract Action Tasks in parallel.
Step 6: Configure Output Destinations (Slack + ClickUp)
Send the summary to Slack and create ClickUp tasks for each action item.
- In Post Slack Summary, set Text to
={{ $json.summary }}and choose a Channel ID. - Credential Required: Connect your slackOAuth2Api credentials.
- In Extract Action Tasks, keep the provided JavaScript to map
action_itemsintotitleanddescription. - Connect Extract Action Tasks to Create ClickUp Task.
- Set List, Team, and Space to your ClickUp IDs (replace
[YOUR_ID]). - Set Name to
={{ $json.title }}and Content to={{ $json.description }}. - Credential Required: Connect your clickUpOAuth2Api credentials.
Step 7: Test and Activate Your Workflow
Validate the full flow from webhook to Slack and ClickUp, then enable it for production.
- Use Incoming Fireflies Hook to trigger a test event (or replay pinned data).
- Confirm Retrieve Transcript returns transcript data and Generate Summary outputs JSON.
- Verify a summary message appears in Slack and tasks are created in ClickUp.
- Toggle the workflow to Active for production use.
Watch Out For
- Fireflies.ai API access can be limited by plan or permissions. If transcript retrieval fails, check your Fireflies dashboard API settings and the meeting ID being sent in the webhook.
- If you’re using AI plus external services, response times vary. If Slack posts before the cleaned JSON is ready, increase wait time or tighten the “Clean AI Output” formatting logic so empty fields don’t slip through.
- ClickUp task creation can fail silently when required fields are missing (like List ID or status rules). Confirm your target Space/List, and make sure the OAuth connection can create tasks in that workspace.
Common Questions
About an hour if your Fireflies, Slack, and ClickUp accounts are ready.
Yes. You won’t write code, but you will connect accounts and paste API credentials. If you can handle basic ClickUp list selection and Slack channel targeting, you’re fine.
Yes. n8n has a free self-hosted option and a free trial on n8n Cloud. Cloud plans start at $20/month for higher volume. You’ll also need to factor in Gemini API usage costs, which depend on how long your transcripts are.
Two options: n8n Cloud (managed, easiest setup) or self-hosting on a VPS. For self-hosting, Hostinger VPS is affordable and handles n8n well. Self-hosting gives you unlimited executions but requires basic server management.
You can. Most customization happens in the AI prompt inside Generate Summary, and in how you format fields in Clean AI Output and Extract Action Tasks. Common tweaks include changing the recap sections (decisions, risks, blockers), mapping owners to a ClickUp assignee rule, and posting different formats to different Slack channels based on meeting name.
Usually it’s expired OAuth credentials or the app not having permission to post in that channel. Reconnect Slack in n8n, then confirm the channel still exists and the bot is allowed to post there.
On n8n Cloud, capacity depends on your plan’s monthly executions, and each meeting typically consumes one run plus a few internal steps. If you self-host, there’s no execution cap, so the real limit becomes your server and AI response time. In practice, most small teams can run this for every meeting without thinking about it. If you start processing lots of long transcripts daily, you’ll want to monitor AI costs and consider queueing or scheduling to avoid bursts.
Often, yes, because this workflow benefits from multi-step formatting, JSON cleanup, and branching logic that gets awkward (and pricey) in simpler tools. n8n also gives you a self-hosted option, which is a big deal if you have lots of meetings or strict data policies. Zapier and Make can still work if you only want “transcript summary to Slack” with minimal structure. But once you want dependable action-item extraction and ClickUp task creation, the extra control matters. If you’re torn, Talk to an automation expert and you’ll get a straight recommendation.
Once this is running, your meetings end with a recap people can skim and tasks that actually exist. Honestly, it’s hard to go back after that.
Need Help Setting This Up?
Our automation experts can build and customize this workflow for your specific needs. Free 15-minute consultation—no commitment required.
