# Uniphore Conversation AI MCP for AI Agents AI Agent Connect > Uniphore Conversation AI connects your AI agent to your meeting data. You can pull transcripts, summaries, and action items from Uniphore directly into your chat. It turns messy calls into searchable data for sales and support teams, making it easy to find specific customer quotes or sentiment trends without manual searching or listening back to old audio files. ## Overview - **Category:** communication-messaging - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_TcPVvPg2b89t9Ygv6B1hi50S5BhjS1cPHrb54Ba5/ai-agent-connect - **Tags:** conversation-intelligence, transcription, sentiment-analysis, meeting-summaries, action-items, analytics ## Description This Connector connects Uniphore conversation data to your AI agent. Instead of jumping between tabs to find a specific transcript or manually copying notes into a spreadsheet, you can just ask your agent for the highlights. It pulls the actual text of the conversation, identifies who said what, and grabs the specific tasks that need to be finished. You can even check the sentiment of the call to see if a customer was happy or frustrated. Because Vinkius makes it easy to connect these types of tools, you can keep your workflow focused on the actual work rather than the data entry. It turns your recorded meetings into a searchable database that your agent can query on the fly. You no longer have to spend your Monday morning listening to hours of audio to find that one specific quote about a competitor or a pricing objection. Your agent can scan the entire history of your calls and surface the exact moment a customer expressed interest in a new feature. This means you can prepare for follow-up meetings with actual facts in hand rather than relying on your memory of a conversation that happened days ago. It moves the needle from just having recordings to actually using the information inside them to drive your sales and support goals. ## Tools ### get_action_items Pulls specific tasks and next steps from a meeting so you don't have to hunt for them in your notes. It helps you stay on top of commitments made during calls. ### get_meeting_analytics Shows you the sentiment, talk ratios, and engagement metrics for any given call. Use this to gauge customer interest quickly. ### list_meetings Lets you see a full list of recorded calls to find the right ID for your query. It's the first step in finding specific data. ### get_meeting_summary Grabs the AI-written summary of a call to get the highlights in seconds. It's perfect for quick catch-ups. ### list_topics Shows every topic and keyword your organization currently tracks. Use this to understand the broad themes in your calls. ### get_meeting Fetches the core details of a specific meeting record. It's the best way to get a quick snapshot of a single conversation. ### search_meetings Finds specific conversations based on topics or keywords you remember. It saves you from scrolling through endless call logs. ### get_transcript Pulls the full, word-for-word transcript of a meeting with speaker tags. You can review exactly what was said. ## Prompt Examples **Prompt:** ``` What were the next steps from the call with Acme Corp yesterday? ``` **Response:** ``` ### Action Items from Acme Corp Call Based on the conversation, here are the next steps: * **Send pricing proposal** (Assigned to: John Doe) * **Schedule technical demo** for next Tuesday * **Follow up on API integration** questions *The customer seemed very interested in the scalability of the platform.* ``` **Prompt:** ``` Give me a summary of the last three support calls regarding the login issue. ``` **Response:** ``` ### Support Call Summaries: Login Issue | Date | Issue Summary | Sentiment | | :--- | :--- | :--- | | Oct 12 | User unable to reset password via email | Frustrated | | Oct 13 | Multi-factor authentication loop | Neutral | | Oct 14 | SSO integration timeout | Frustrated | **Key Takeaway:** Customers are primarily struggling with the MFA loop during SSO setup. ``` **Prompt:** ``` What was the customer's sentiment in the meeting about the new pricing? ``` **Response:** ``` ### Pricing Sentiment Analysis The customer expressed **mixed sentiment** regarding the new pricing: * **Positive:** They appreciated the move to a per-user model. * **Negative:** They were concerned about the 15% increase for enterprise-level seats. **Talk Ratio:** The customer spoke for 65% of the meeting during the pricing discussion. ``` ## Capabilities ### Pull speaker-tagged transcripts Get the full word-for-word text of a meeting with clear labels for every speaker. ### Get AI-generated summaries Retrieve a concise overview of the key discussion points from any recorded call. ### Extract action items Pull specific tasks and next steps identified during a conversation automatically. ### View conversation analytics See talk ratios, sentiment scores, and engagement metrics for your meetings. ### Search past meetings Find specific conversations by searching for keywords or topics discussed. ### List organization topics See every topic and keyword your company currently tracks across conversations. ## Use Cases ### Finding specific pricing objections A sales rep needs to know what a prospect said about pricing in a call from three weeks ago. They ask their agent to search meetings for 'pricing' and pull the transcript. ### Auditing support sentiment A support lead wants to see if customers are complaining about a new bug. They use `get_meeting_analytics` to check sentiment trends on recent calls. ### Prepping for an account check-in An account manager needs to prep for a check-in. They ask the agent for the `get_meeting_summary` of the last three interactions to get up to speed. ### Compliance keyword auditing An ops manager needs to find every time 'refund' was mentioned. They use `search_meetings` to find all relevant calls and pull the transcripts. ## Benefits - Stop manual note-taking by using `get_meeting_summary` to get a high-level overview of every call instantly. This saves you from having to listen back to recordings just to find the main points of a long discovery call. - Identify high-priority leads faster by using `get_action_items` to see exactly what the customer asked for. You'll know exactly what to do next without digging through messy notes. - Spot customer frustration early by pulling sentiment data through `get_meeting_analytics` during routine audits. It helps you catch issues before they turn into churn. - Find specific information in seconds by using `search_meetings` instead of scrolling through old call logs. It makes your historical data actually useful for daily work. - Keep your team aligned on goals by using `get_transcript` to review exactly what was promised during a closing call. It ensures everyone is on the same page regarding next steps. - Track organizational trends over time by using `list_topics` to see which themes are appearing most often. This helps you see the big picture of customer needs. ## How It Works The bottom line is you get instant access to the data hidden inside your recorded conversations. 1. Subscribe to this Connector and provide your Uniphore Access Token and Organization ID. 2. Connect your preferred AI client like Claude, Cursor, or Windsurf. 3. Ask your agent to find specific meeting details, pull transcripts, or summarize recent calls. ## Frequently Asked Questions **Can Uniphore Conversation AI help me find specific keywords in old calls?** Yes. You can ask your agent to search for any keyword or topic discussed in your history. It will find the relevant meetings and pull the specific transcripts or summaries for you. **How does Uniphore Conversation AI handle meeting summaries?** It pulls the AI-generated summaries directly from your Uniphore account. Your agent can then present these highlights to you in plain text or as part of a larger report. **Can I use Uniphore Conversation AI to see how long people talked?** Yes, it provides talk ratios as part of the conversation analytics. This helps you see the balance of the conversation between your team and the customer. **Does Uniphore Conversation AI work for sales teams?** It's a great fit for sales teams. It helps you quickly review call transcripts, identify action items, and track customer sentiment across your entire pipeline. **Can I get the full transcript of a call using Uniphore Conversation AI?** Absolutely. Your agent can pull the complete, speaker-tagged transcript of any recorded call so you can review the exact dialogue. **How do I find action items from my Uniphore meetings?** You can simply ask your agent to list the action items for a specific meeting. It will pull the tasks and next steps identified by the Uniphore AI. **How do I get my Uniphore Access Token and Org ID?** Log in to your Uniphore account, navigate to Settings > API Access, and generate a new token. Your Organization ID is displayed in the same settings panel. **Can I get AI-generated summaries of meetings?** Yes! Use the get_meeting_summary tool with a meeting ID to retrieve the concise AI summary of that call. **Can I extract action items from a call?** Yes! Use the get_action_items tool with a meeting ID. Uniphore AI identifies next steps and assigns them to speakers if possible. **Can I search for specific topics in past meetings?** Yes! Use the search_meetings tool with relevant keywords or topic names to find matching recorded calls.