Attention is a sales conversation intelligence and coaching platform that records, transcribes, and analyzes customer-facing meetings. The Attention connector lets your Lovable app read conversations, transcripts, and scorecards from Attention, and write back snippets, scorecard results, and conversation updates, so you can build dashboards, briefings, coaching tools, and QA workflows on top of your sales motion. With Attention, your app can:Documentation Index
Fetch the complete documentation index at: https://docs.lovable.dev/llms.txt
Use this file to discover all available pages before exploring further.
- List and retrieve sales conversations and meeting metadata
- Read full meeting transcripts, participants, and timing details
- Surface scorecards, insights, and coaching signals from the Attention API
- Update, archive, or import conversations, and create snippets and scorecard results
- Ask natural-language questions across your conversation history with Attention’s semantic AI endpoint
- Pull and manage organization, team, and user data
- Combine conversation data with the rest of your app’s content (CRM views, internal tools, AI summaries)
Common use cases and example apps
| Example app | Example prompt | Description |
|---|---|---|
| Deal-room conversation viewer | Use Attention and build a deal-room view that lists recent conversations for an account, with transcripts, participants, and durations. | Bring meeting context next to the deal. The app fetches conversations from Attention, filters by account or owner, and renders transcripts and participant lists alongside whatever deal info your team already tracks. |
| Weekly sales briefing app | Use Attention and build a weekly briefing that summarizes last week’s customer calls and emails it to the team. | Turn raw conversations into a leadership-ready digest. The app pulls the last seven days of conversations, summarizes themes and risks, and produces a recurring digest the team can scan in minutes. |
| Coaching and scorecard dashboard | Use Attention and build a coaching dashboard that shows scorecard results and trends per rep. | Make coaching signals legible across the team. The app reads scorecards and conversation metadata from Attention and presents trends per rep, so managers can spot what to coach on next. |
| Ask-your-calls chatbot | Use Attention and build a chatbot that answers questions about our sales calls, like “what objections came up about pricing last month?” | Query meetings in plain English The app uses Attention’s Ask Attention endpoint to run semantic queries across conversations and returns grounded answers with source calls. |
| Auto-tagging and QA tool | Use Attention and build a tool that submits scorecard results and updates conversations after an external QA review. | Push structured QA back into Attention. The app reads conversations, lets reviewers score them in a custom UI, then writes scorecard results and updates conversation metadata back via the Attention API. |
| Onboarding library for new reps | Use Attention and build an onboarding library of best-of calls grouped by deal stage. | Curate calls into a learning experience. The app pulls selected conversations from Attention and organizes them by stage or topic so new reps can ramp by listening to real customer calls. |
| Risk and pipeline review app | Use Attention and build a pipeline review that flags deals where recent calls mention risk language. | Surface risk signals from conversations. The app combines conversation transcripts and scorecards to highlight deals that need attention before the next forecast call. |
| Internal “voice of the customer” feed | Use Attention and build an internal feed of customer quotes pulled from recent meetings. | Distribute the voice of the customer across the company. The app surfaces highlights from Attention conversations as a feed product, support, and design teams can browse. |
How to connect Attention
Workspace admins and owners can connect Attention. You can create multiple Attention connections using different API keys. This is useful for separating environments (for example, development and production), isolating access for specific apps, or managing usage limits independently. When the connection is created, Attention can be linked to any project in the workspace. Multiple projects can use the same connection.Prerequisites
Before connecting Attention, make sure you have:- An Attention account with admin role permissions in your organization
- An Attention API key
- Lovable workspace owner or admin role
All API requests made through this connector use your Attention API key. Usage counts toward your Attention plan limits and billing is handled directly by Attention, not Lovable.
Step 1: Create an Attention API key
An Attention API key is an organization-level Bearer token that lets your Lovable app authenticate against Attention’s v2 API. You must have Admin role permissions in Attention to create and manage organization-level API keys. To create an Attention API key:Log in to Attention
Go to app.attention.tech and sign in with Google, SSO, or your email and password.
Open settings
Click your profile avatar in the top-left corner and select Settings from the dropdown menu.
Create a new API key
Click + Create API Key in the top-right corner and give the key a descriptive name, for example
Lovable integration.Your API key functions like a password. Keep it secure and never share it publicly. You’ll need it in the next step.
Step 2: Connect Attention to Lovable
You can create multiple connections using different API keys. To connect Attention:Configure the connection
- Display name: Name the connection, for example
Attention Prod. - API key: Paste your Attention API key.
- Who can access this connection: Keep access limited to specific people or invite the entire workspace. See Connection-level access for more information.
Working with conversation data
The Attention API returns rich conversation objects that include metadata, participants, and full transcripts, and supports updating, archiving, and importing conversations. When prompting Lovable, be specific about what you want. Conversations include sensitive customer information. Plan how the data is displayed in your app and who can see it before sharing the app broadly inside your company.Limitations
The Attention connector cannot:- Receive Attention webhooks or real-time conversation events (poll the API on a schedule instead)
- Support per-end-user Attention login (each end user connecting their own Attention account). All requests use the workspace API key, so every end user of your Lovable app sees the same Attention data.
429 Too Many Requests responses and add retry-with-backoff in your edge functions.
If a request fails because the API key has been revoked or rotated in Attention, you’ll need to generate a new key and reconnect. There is no automatic refresh for API keys.
How to unlink projects from a connection
Editors and above can remove specific projects from a connection without deleting the connection entirely. The connection will remain available for other projects. To unlink projects:
When unlinked, those projects will no longer have access to through this connection. If a project needs again, you can link it to any available connection.