> ## Documentation Index
> Fetch the complete documentation index at: https://docs.groovin.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Groovin MCP Use Cases: ready-to-use prompts

> Practical AI prompt examples using Groovin MCP to automate LinkedIn prospecting, CRM sync, follow-ups, and lead discovery for sales teams.

Groovin MCP works best when you give your AI assistant clear, specific instructions that combine LinkedIn context with CRM data. Rather than vague requests, the prompts that get the best results are the ones that tell your assistant exactly what data to fetch, how to cross-reference it, and what action to take next. This page shows proven prompt patterns and use cases you can adapt to your workflow.

## Writing effective prompts

> [!TIP]
>
> Always start your prompt with **"Use Groovin to..."** so your AI assistant knows to activate the Groovin MCP tools rather than guessing which connector to use.

* **Be specific about date ranges** — say "since May 1st 2025" rather than "recently" so Groovin fetches exactly the data you need.
* **Chain actions in sequence** — first fetch data, then cross-reference the CRM, then take action. Breaking the task into steps gives the assistant a clear execution path.
* **Use follow-up prompts to refine** — ask for a draft first, then ask the assistant to adjust the tone or shorten it. Iterating in steps produces better output than one overly complex prompt.
* **Combine Groovin with your CRM's MCP for full context** — for example, use Groovin MCP to pull LinkedIn data and your CRM's MCP (Attio, HubSpot or Affinity) to pull deal stage, last meeting notes, and email history in the same conversation.

## Use case examples

> [accordion] Import LinkedIn conversations and add missing contacts
>
> **Prompt:**
>
> ```text theme={null}
> Use Groovin to list all my LinkedIn conversation attendees with activity since May 1st 2025.
> Check in bulk which ones are already in the CRM.
> Create CRM contacts for the ones that aren't, and enable LinkedIn message sync.
> ```
>
> **What happens:** Groovin fetches your conversation attendees with `list_linkedin_conversation_attendees`, calls `lookup_crm_contacts` in a single batch to check who already exists in your CRM, and creates the missing ones with `create_crm_contacts` (with `sync_messages` on). At the end you have a fully populated CRM with LinkedIn sync active for every conversation.

> [accordion] Find and draft follow-ups for stale conversations
>
> **Prompt:**
>
> ```text theme={null}
> Use Groovin to find LinkedIn conversations where I sent the last message
> more than 4 days ago and haven't received a reply.
> For each, get the person's CRM record and any related deal context.
> Then draft a short, specific follow-up message.
> ```
>
> **What happens:** Groovin fetches your messages and filters them by direction and date to surface only the conversations that have gone cold. It then calls `lookup_crm_contacts` for each person to pull their CRM record and deal context. The AI uses all of that to draft a personalised follow-up for each conversation.
>
> > [!TIP]
> >
> > Combine with the Attio, HubSpot or Affinity MCP to pull deal stage, last meeting notes, and email history for even richer follow-up context.

> [accordion] Generate a morning briefing from your LinkedIn inbox
>
> **Prompt:**
>
> ```text theme={null}
> Use Groovin to scan my LinkedIn inbox from the last 24 hours.
> Flag the 3 most important conversations I should reply to today
> and draft a response for each.
> ```
>
> **What happens:** Groovin fetches all messages received in the last 24 hours. The AI ranks them by relevance — prioritising new leads, open deals, and time-sensitive asks — and drafts a contextual reply for each of the top conversations. Run this every morning to start the day with a clear action list.

> [accordion] Surface new connections that aren't in your CRM
>
> **Prompt:**
>
> ```text theme={null}
> Use Groovin to find new connections I've had over the past 3 months
> that I haven't contacted yet.
> Check if they're in the CRM and use their LinkedIn profile data
> to check if they match my ICP.
> ```
>
> **What happens:** Groovin calls `list_linkedin_relations` with a `connected_after` filter to surface new connections, and runs `lookup_crm_contacts` in bulk to see who is already in your CRM. For a small shortlist of unpaired matches, it can call `get_linkedin_profile` sparingly (stay around 100 profile views a day) to check ICP fit. You end up with a shortlist of qualified leads hiding in plain sight in your network.

> [accordion] Track and escalate pending connection requests
>
> **Prompt:**
>
> ```text theme={null}
> Use Groovin to find connection requests I sent last week that are still pending,
> then send a Slack message to my CEO / Managing Partner
> with the list of LinkedIn profiles to invite.
> ```
>
> **What happens:** Groovin calls `list_linkedin_pending_invitations` with a `sent_after` filter to find all unaccepted invites from the specified period. The AI then formats the results as a ready-to-paste Slack message so you can escalate directly to your CEO or another team member who can follow up through a different channel.

> [accordion] Turn prompts into recurring tasks with Claude
>
> Once you've validated a prompt pattern and it reliably returns the output you want, you can turn it into a scheduled task. Using Claude's Projects feature, Claude Cowork, or Claude Code, describe the frequency you want (e.g. "run every weekday at 8am") and save the prompt as a recurring instruction. Your AI assistant will execute the full Groovin workflow automatically, without you having to re-enter the prompt each time.

> [card] [Prompt Library — ready to copy-paste](https://docs.groovin.ai/mcp/prompt-library.md)
>
> Ready-made prompts you can drop straight into your AI assistant.

> [card] [Skills — write in your own voice](https://docs.groovin.ai/mcp/skills.md)
>
> Installable skills that learn how you write on LinkedIn, then draft and send messages that sound like you.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.