# MonkeyLearn MCP for AI Agents AI Agent Connect > MonkeyLearn lets you run complex NLP tasks like sentiment analysis and entity extraction through your AI agent. Instead of writing Python scripts or manual labeling, you can just ask your agent to process thousands of customer reviews, categorize support tickets, or pull specific data points from messy text. It connects your existing MonkeyLearn models directly to your workflow. ## Overview - **Category:** customer-support - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_cjNrQPTPb2dGyXsgNWrkuHpvGkEq1lXSllJz11sl/ai-agent-connect - **Tags:** text-classification, entity-extraction, sentiment-analysis, nlp, machine-learning, data-labeling ## Description Connect your account to any AI agent and run NLP text analysis through natural conversation. Imagine you have a mountain of customer feedback from different sources. Usually, you'd have to export that data, upload it to a platform, and then spend hours looking at the results or building custom scripts to sort through it. This Connector changes that by letting you talk to your data directly. You can ask your agent to take a batch of reviews and tell you which ones are actually angry versus just being disappointed. It handles the heavy lifting of running your machine learning models so you can stay focused on what the data actually means for your business. Because Vinkius hosts this connection, you don't have to worry about the plumbing. You just get a direct line between your conversation and your models. It turns a multi-step data science project into a simple chat. Instead of manually tagging hundreds of support tickets or trying to write complex patterns to pull names out of emails, you can just describe what you need. Your agent does the work, and you get back organized, structured results ready for your next step. You can quickly check your model status or run entire workflows with a single prompt, making it a massive time saver for anyone dealing with large volumes of unstructured text. ## Tools ### list_classifier_tags See all the tags your models use for categorization. This helps you understand how your data is currently being organized. ### list_classifiers Get a full list of your available text classifiers. You can quickly see which models are ready to use for your next task. ### list_extractor_tags View the labels used by your text extractors. This is useful for knowing what specific data points your models can pull out. ### list_extractors See all the entity extraction models in your account. It lets you know which tools are available for pulling out names or dates. ### list_model_versions Check which versions of your models are currently active. This helps you ensure you aren't using outdated logic for your analysis. ### list_nlp_workflows See all the multi-step workflows you've built. You can quickly identify which complex pipelines are available to run. ### run_workflow Execute a specific multi-step NLP pipeline. This is the best way to handle complex, multi-stage text processing in one go. ### classify_text Send text to a model to get a sentiment or category tag. It's the fastest way to organize large amounts of unstructured feedback. ### extract_text_entities Pull structured data like names, dates, and locations from a block of text. It turns messy paragraphs into clean, usable data points. ### get_api_status Check if your connection to MonkeyLearn is active. Use this to quickly troubleshoot any issues with your integration. ### get_classifier_details Get deep info on a specific text classifier. This is helpful when you need to know the exact specs of a model you're using. ### get_extractor_details Get deep info on a specific text extractor. Use this to see the specific details of how your entity extraction is set up. ## Prompt Examples **Prompt:** ``` Can you look at these 5 reviews and tell me which ones are from angry customers? ``` **Response:** ``` ### Sentiment Analysis Results | Review | Sentiment | Confidence | | :--- | :--- | :--- | | 'The product is amazing but delivery was slow.' | **Mixed** | 0.82 | | 'I hate the new interface, it is so confusing.' | **Negative** | 0.98 | | 'Best purchase I've made all year!' | **Positive** | 0.99 | | 'The app crashes every time I open it.' | **Negative** | 0.95 | | 'It's okay, does what it says.' | **Neutral** | 0.70 | **Summary:** 2 reviews are clearly negative, primarily focusing on UI confusion and app stability. ``` **Prompt:** ``` Extract the names and locations from this list of notes: 'Met with Sarah from Acme in Chicago on Tuesday'. ``` **Response:** ``` ### Extracted Entities * **Person:** Sarah * **Organization:** Acme * **Location:** Chicago * **Date:** Tuesday **Total entities found:** 4 ``` **Prompt:** ``` What workflows do I have available to run? ``` **Response:** ``` ### Available NLP Workflows * **Customer Feedback Loop:** Analyzes sentiment and extracts keywords in one pass. * **Lead Scraper:** Extracts names and contact info from raw notes. * **Ticket Router:** Categorizes support tickets by intent and priority. ``` ## Capabilities ### Classify text by sentiment or intent Your agent can automatically tag large batches of text with specific labels like 'angry' or 'billing issue'. ### Pull names and addresses from messy notes Extract structured details like people, locations, and dates from unstructured blocks of text. ### Run multi-step NLP pipelines Trigger complex, multi-stage text processing workflows with a single natural language command. ### List your available classifiers See all the text classification models currently available in your account. ### Check your connection status Verify that your account is correctly connected and ready for analysis. ## Use Cases ### Sentiment Analysis A product manager asks the agent to find all negative reviews from last week and summarize the main complaints. ### Lead Extraction A sales rep wants to pull names and company names out of a list of raw outreach notes to create a CRM list. ### Ticket Routing A support lead asks the agent to classify incoming emails by intent so the team knows which tickets need immediate attention. ### Keyword Research A marketer wants to extract the most common product mentions from a batch of survey results to plan a new campaign. ## Benefits - Skip the code: Use classify_text to get labels on data without writing a single line of Python or managing complex API calls. - Automate pipelines: Use run_workflow to trigger complex NLP sequences that usually require manual intervention or custom scripts. - Clean your data: Use extract_text_entities to turn messy customer emails into structured lists of names, dates, and locations. - Manage models easily: Use list_classifiers and list_model_versions to keep track of your ML assets without leaving your chat interface. - Fast troubleshooting: Use get_api_status to verify your connection instantly, so you spend less time debugging and more time analyzing. ## How It Works The bottom line is you get to run professional NLP models using simple natural language commands. 1. Connect your MonkeyLearn API key in your client. 2. Tell your agent which model or workflow to use. 3. Receive structured results like tags or extracted entities. ## Frequently Asked Questions **Does MonkeyLearn MCP help with sentiment analysis?** Yes, it lets your agent use your MonkeyLearn models to instantly tag text as positive, negative, or neutral. **Can I use MonkeyLearn MCP to extract names from text?** Yes, you can use the entity extraction tools to pull out specific details like names, dates, and locations from messy notes. **How does MonkeyLearn MCP connect to my existing account?** It uses your MonkeyLearn API key to give your agent direct access to your custom models and workflows. **Can MonkeyLearn MCP handle multi-step NLP tasks?** Yes, it can trigger complex workflows that perform multiple operations on a piece of text in one go. **Is MonkeyLearn MCP good for marketing research?** It's great for marketers who need to quickly pull keywords or common themes out of large sets of survey responses. **Can I see my models using MonkeyLearn MCP?** Yes, your agent can list all your classifiers, extractors, and model versions so you know what's available. **Can I classify text by sentiment or topic?** Yes. Point to any classifier model ID and pass text to get classification results with confidence scores. **How does MonkeyLearn authentication work?** MonkeyLearn uses `Authorization: Token {API_KEY}` header against `api.monkeylearn.com/v3`. **Can I extract named entities from text?** Yes. Use an extractor model to pull keywords, people names, organizations, locations, and more from raw text.