# Uniphore Conversation AI MCP > Uniphore Conversation AI accesses meeting transcripts, summaries, and analytics data via Uniphore API. This server lets your agent pull specific structured info—like action items, key decisions, or sentiment analysis—from recorded calls without needing a human to review the audio first. ## Overview - **Category:** communication-messaging - **Price:** Free - **Tags:** conversation-intelligence, transcription, sentiment-analysis, meeting-summaries, action-items, analytics ## Description ### Uniphore Conversation AI: Meeting Insights MCP Server Your agent handles complex call intelligence. You can treat recordings like a structured database, pulling out specific data points using natural language prompts—no human has to listen to the audio first. ### Finding Your Data To start, you'll need to know what data exists. Use `list_meetings` to pull up every recorded conversation and call ID available in your organization. If you know roughly when or who was involved, you can narrow things down fast by using `search_meetings`, which filters past records based on keywords, specific topics, or a date range. You don't even need to guess what kind of data is available; run `list_topics` and you'll get an index of every keyword or topic tracked across all recordings. When you know the exact meeting ID, you can get basic details about that single call using `get_meeting`. ### Raw Content Extraction When you need to see everything that was said, run `get_transcript`. This outputs the full, speaker-tagged text transcript for an entire recording, so every word spoken is accounted for. For a quick read of what actually happened, use `get_meeting_summary` to get a concise, AI-written overview of all the main topics discussed during that call. ### Deep Dive Analytics and Action Items This server pulls out structure from unstructured audio. It doesn't just give you text; it gives you actionable data points. To pull out next steps—the tasks and deliverables—run `get_action_items`. This tool analyzes the meeting content specifically to isolate assignees, specific tasks, and due dates. You'll get a clean list of everything that needs doing. If you need to know how well people talked or what the general vibe was, use `get_meeting_analytics`. This generates detailed conversation metrics for you, including sentiment analysis (telling you if the call was positive, negative, or neutral), talk ratios between participants, and a breakdown of primary topics discussed. It’s a full performance audit of the conversation. ### How it Works with Your Agent Your agent manages this flow. You point your AI client at the server, give it the necessary access credentials for Uniphore, and then ask direct questions about your recorded calls. The agent executes the tool—say, `get_action_items` on a specific meeting ID—and brings back structured JSON data. This means you don't have to parse through walls of text; you just get the list: 'Task A assigned to John by Friday,' or 'Sentiment was negative during the budget discussion.' You’ll find it makes reviewing massive amounts of recorded calls fast and precise. ## Tools ### get_action_items Pulls a list of specific tasks and next steps identified during a meeting. ### get_meeting_analytics Generates detailed metrics on the conversation, including sentiment and talk ratios. ### get_meeting Retrieves general details about a single, known recorded meeting or call. ### list_meetings Lists all available recorded meetings and calls so you can find a specific ID. ### search_meetings Filters and finds specific past meetings by searching for a keyword, topic, or date range. ### get_meeting_summary Gets a concise summary of all key topics discussed during the meeting. ### list_topics Displays all keywords or topics that have been tracked across the organization's recordings. ### get_transcript Outputs the full, speaker-tagged text transcript for an entire recording. ## Prompt Examples **Prompt:** ``` Show me the summary for meeting MTG-123. ``` **Response:** ``` Meeting Summary: Discussed Q4 targets, agreed on new pricing strategy, and scheduled follow-up for next week. ``` **Prompt:** ``` Get the transcript for meeting MTG-456. ``` **Response:** ``` Transcript: - Agent: Thanks for joining. What can I help with today? - Customer: I have a question about my recent invoice. ``` **Prompt:** ``` What are the action items from the last sales call? ``` **Response:** ``` Action Items: 1. Send pricing proposal to client (Assignee: John Doe) 2. Schedule technical demo for next Tuesday (Assignee: Jane Smith) ``` ## Capabilities ### Retrieve full call transcripts Gets a speaker-tagged record of every word spoken during a specific meeting. ### Extract key summaries and insights Generates concise, AI-written overviews of the main discussion points from any recorded call. ### Identify next steps and tasks Analyzes meeting content specifically to pull out actionable items, including assignees and due dates. ### Analyze performance metrics Provides conversation analytics like sentiment trends, talk ratios, and primary topics discussed. ### Find specific meeting data Allows you to list or search for past meetings using keywords, topics, or date ranges. ## Use Cases ### Onboarding a new sales team member The manager needs to train a rep on handling objections related to pricing. Instead of pairing the rep up for roleplay, they run `search_meetings` across all recorded calls using keywords like 'pricing' or 'cost'. The agent then uses `get_transcript` and filters it down, providing the new hire with 10 minutes of targeted training material. ### Identifying process bottlenecks in support The Support Manager notices high call volume but doesn't know why. They use `get_meeting_analytics` on all tickets from last week. The results show a spike in 'billing confusion' sentiment, pointing the team directly to the outdated billing portal as the root problem. ### Following up after an executive meeting The project lead attends a critical planning session. Instead of taking manual notes that get lost, they immediately prompt their agent: 'What are the action items from MTG-XYZ?'. The agent runs `get_action_items`, delivering a clean list to the team chat within seconds. ### Researching product feedback across multiple regions A Product Manager wants to know how different international clients feel about Feature X. They use `list_topics` to confirm 'Feature X' is tracked, then run `search_meetings` combining the topic with regional identifiers (e.g., 'Germany'). The agent returns all relevant transcripts for review. ## Benefits - **Stop watching video files.** Instead of spending time scrubbing through hours of audio, you can ask the agent to run `get_meeting_summary` or `get_action_items`, and it hands you a bulleted list of decisions made. This saves hours per week. - **Track customer sentiment automatically.** The `get_meeting_analytics` tool lets you analyze conversation metrics across hundreds of calls at once. You'll see if your support agents are struggling with the same topic, or if a product change is causing negative buzz. - **Never lose an action item again.** Instead of relying on shaky human notes after a call, use `get_action_items` to pull out structured tasks immediately. It tells you who owns the task and when it's due. - **Audit compliance effortlessly.** Need proof that certain keywords were discussed? Use `search_meetings` with specific topics or date ranges. You can quickly find all calls mentioning 'NDA' or 'pricing change' for a full audit trail. - **Get raw data instantly.** If the AI summary misses something, you can always fall back to `get_transcript`. This gives your agent the entire, speaker-tagged text record—the source of truth—in seconds. ## How It Works The bottom line is that you treat recorded meetings like a structured API endpoint, pulling out specific intelligence instead of reading through audio files. 1. Subscribe to this server and enter your Uniphore Access Token and Organization ID. 2. Ask your AI client to locate a meeting record (using `list_meetings` or `search_meetings`). 3. Tell the agent what data you need—e.g., 'Give me the summary' or 'What are the action items?'—and it runs the appropriate tool. ## Frequently Asked Questions **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. **When I use the `get_transcript` tool, what format do I receive for the meeting dialogue?** You get speaker-tagged transcripts that show who said what and exactly when. This raw data separates agent speech from customer speech, making it easy to track participation balance. **What kind of metrics does `get_meeting_analytics` return for a call?** It returns specific conversation analytics like talk ratios, sentiment scores, and topic engagement. These numbers help you measure emotional tone and who was driving the discussion. **How do I find out which meetings are available before calling `get_meeting`?** Run the `list_meetings` tool first. This call provides a comprehensive list of all recorded meeting IDs, allowing you to select the exact data point you need. **How can I view the catalog of keywords and topics tracked by Uniphore Conversation AI?** Use `list_topics` to pull a complete catalog of every keyword tracked in your organization. This lets you confirm that your searches are focused on relevant, monitored domains.