# Writer MCP for AI Agents AI Agent Connect > Writer (AI Enterprise LLM) MCP gives your AI agent direct access to Writer's enterprise models, Knowledge Graphs, and document management tools. It lets you build RAG workflows, manage large-scale file pipelines, and run no-code applications through a single connection. It's built for teams needing secure, production-ready generative AI. ## Overview - **Category:** developer-tools - **Price:** Free - **Endpoint:** https://edge.vinkius.com/undefined/ai-agent-connect - **Tags:** llm, enterprise-ai, knowledge-graph, rag, palmyra ## Description This Connector connects your AI agent to Writer's enterprise-grade platform. It lets you pull data from Knowledge Graphs to power Retrieval-Augmented Generation (RAG) workflows without the usual headache of manual data plumbing. You can manage your entire document library, upload files for model context, and even run no-code applications to automate complex tasks. It handles the heavy lifting of enterprise document processing so you don't have to worry about data silos. You can find this in the Vinkius catalog to get your agent connected to these high-performance models and your internal knowledge bases in minutes. It's about giving your agent the same tools a human would use to manage a corporate knowledge base, but at the speed of an LLM. ## Tools ### download_file Get the binary content of a specific file. Use this when you need to pull raw data. ### analyze_vision Use a prompt to analyze images or documents. This lets your agent see and describe visual data. ### ask_question Query one or more Knowledge Graphs for an answer. This is the primary way to perform RAG. ### chat_completion Generate a chat response using Writer's enterprise models. This is perfect for building conversational interfaces. ### generate_application_content_async Start a long-running no-code application job. This returns a job ID for you to check later. ### generate_application_content Run a no-code application job synchronously. Use this for quick results from your pre-built apps. ### get_application_job Check the status and results of a specific application job. This is how you track async tasks. ### get_application See the configuration and inputs for a specific application. Use this to understand how an app is set up. ### get_file Get the metadata for a specific file. This is useful for organizing your document library. ### list_application_jobs See a list of all jobs for a specific application. Use this to monitor high-volume task runs. ### list_applications See all your deployed no-code agents or applications. This helps you keep track of your automated workflows. ### list_files Get a paginated list of all your uploaded files. Use this to audit your current storage. ### list_graphs See all the Knowledge Graphs you've created. This helps you manage your RAG data sources. ### list_models See which Writer models are currently available. This ensures you are using the latest enterprise tech. ### remove_file_from_graph Take a specific file out of a Knowledge Graph. Use this to update your RAG context without deleting the file. ### retry_application_job Run a failed application job again. This helps you recover from temporary errors in your workflows. ### text_completion Generate a text completion for a single prompt. Use this for straightforward, one-off text generation. ### translate_text Translate text between different supported languages. This is great for global content teams. ### upload_file Upload a file to use for Knowledge Graphs or vision tasks. This is the first step in building your RAG context. ### web_search Search the internet for a specific query. Use this to give your agent access to real-time information. ### create_graph Build a new Knowledge Graph from scratch. Use this to organize your data for RAG. ### delete_file Permanently remove a file from your storage. This helps keep your graph context clean. ### parse_pdf Convert a PDF into text or markdown format. This is a legacy tool for extracting content from documents. ### add_file_to_graph Attach an uploaded file to a specific Knowledge Graph. This connects your documents to your RAG system. ## Prompt Examples **Prompt:** ``` Create a Knowledge Graph for our 2024 marketing strategy and add a file called 'goals.pdf' to it. ``` **Response:** ``` I've created the '2024 Marketing Strategy' Knowledge Graph for you. I have also successfully attached 'goals.pdf' to it. You can now ask me questions about the strategy goals or ask me to summarize specific sections of the document. ``` **Prompt:** ``` What does our current policy say about remote work in the UK? ``` **Response:** ``` Based on the 'HR Policy' Knowledge Graph, the current policy for UK-based employees is as follows: * **Eligibility**: All full-time employees are eligible. * **Days per week**: Up to 3 days per week from a home office. * **Equipment**: The company provides a standard home-office stipend. Would you like me to look up the specific requirements for the US office as well? ``` **Prompt:** ``` I need to generate 10 blog post outlines based on our latest product launch. Can you start that now? ``` **Response:** ``` I've started an asynchronous job to generate 10 blog post outlines based on the 'Product Launch 2024' documentation. **Job Details:** * **Job ID**: `app_job_88291` * **Status**: In Progress I'll let you know as soon as the outlines are ready, or you can ask me to check the status of job `app_job_88291` at any time. ``` ## Capabilities ### Generate enterprise text completions Use high-performance models to create chat responses and long-form content. ### Manage Knowledge Graphs Build and organize data structures to power Retrieval-Augmented Generation. ### Handle document pipelines Upload, list, and delete files to maintain your model's context. ### Run no-code applications Trigger and monitor automated jobs for complex, multi-step tasks. ### Search the web Pull in real-time information to keep your agent's knowledge current. ### Analyze visual data Process images and documents using vision-capable model prompts. ## Use Cases ### RAG for internal HR policies An HR manager asks the agent about a specific policy. The agent queries a Knowledge Graph to find the exact answer in a massive document library. ### Automated content production A content lead wants to generate 50 product descriptions. The agent starts an asynchronous no-code application job to handle them in bulk. ### Web-enhanced market research A researcher needs the latest news on a topic. The agent performs a web search to pull in current data before summarizing it for the team. ### Bulk document processing A legal team needs to extract info from 100 PDFs. The agent uploads the files and uses vision analysis to summarize each one. ## Benefits - Build RAG workflows faster by using structured graphs to query your internal data rather than relying on raw file searches. - Manage large document libraries easily by uploading and listing files to keep your model context organized and clean. - Run complex automated tasks by triggering no-code applications to handle high-volume workflows in the background. - Get real-time info by searching the web to ensure your agent isn't stuck with outdated training data during research. - Handle multi-language content by using translation tools to reach a global audience quickly and accurately. - Analyze visual data by processing images and documents in one step using vision-capable model prompts. ## How It Works The bottom line is you get a bridge between your internal data and enterprise-grade AI models. 1. Subscribe to the Writer MCP and grab your API key from the Writer dashboard. 2. Plug the key into your AI client and connect the Connector. 3. Start asking your agent to query Knowledge Graphs or generate content. ## Frequently Asked Questions **How does the Writer MCP help with RAG?** It provides tools to create Knowledge Graphs and ask questions against them, making it easy to build Retrieval-Augmented Generation systems. **Can I use this Connector to manage my company's documents?** Yes, you can upload, list, and delete files to keep your model context and Knowledge Graphs organized. **Does the Writer MCP support web searching?** Yes, it includes a web_search tool so your agent can pull in real-time information from the internet. **Can I run long-running tasks with the Writer MCP?** You can use the async application tools to start a job, get a job ID, and check the status later. **Is this Connector good for multi-language content?** Yes, it includes a translation tool that lets your agent convert text between supported languages quickly. **How do I use the Writer MCP for image analysis?** You can use the vision tool to ask your agent questions about images or documents you've uploaded. **How do I see which Palmyra models are available in my account?** Use the `list_models` tool. It will return a list of all active models in your Writer account, such as palmyra-x5 or palmyra-med. **Can I use my own documents for AI responses?** Yes. First, use `upload_file` to add your documents, then `create_graph` and `add_file_to_graph` to build a Knowledge Graph for RAG-based querying. **Does this support conversational history?** Yes, the `chat_completion` tool accepts a `messages` array, allowing you to maintain full context for multi-turn dialogues with the model.