# Prismatic MCP for AI Agents AI Agent Connect > Prismatic lets you manage your embedded iPaaS environment through an AI client. You can create customer records, deploy integration instances, and pull execution logs without leaving your workspace. It turns your Prismatic environment into a conversational interface for handling B2B integration lifecycles. ## Overview - **Category:** developer-tools - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_Jp8xQVyawqmk1kh1SNDCvDluadVzruFklDLm4lQl/ai-agent-connect - **Tags:** embedded-ipaas, integration-platform, workflow-automation, api-management, saas-integrations ## Description Managing an embedded iPaaS usually involves a lot of tedious context switching. You have to keep your terminal open, your browser tabs organized, and your API keys handy just to perform basic tasks like checking a customer's status or pulling logs for a failed workflow. It's a friction-heavy process that slows down your engineering team and keeps your support staff busy with manual lookups. This Connector changes that by letting your AI agent handle the heavy lifting. Instead of navigating a complex dashboard, you can just tell your agent to set up a new integration instance or pull the logs for a specific customer's failed workflow. You'll be able to manage the entire lifecycle of your B2B integrations using plain language. Because Vinkius hosts this Connector, you get a reliable way to connect your Prismatic API token to your favorite tools like Cursor or Claude. It's about moving from manual configuration to conversational management, whether you're debugging a production issue or onboarding a new batch of users. You can quickly verify which integrations are active, create new customer records with custom IDs for better mapping, and push configuration changes live to your production environment. It turns a series of clicks into a conversation, letting you focus on building better products rather than managing the plumbing of your integration platform. ## Tools ### create_customer Adds a new customer record to your platform with a custom external ID for better mapping. ### create_instance Generates a new integration instance for a specific customer account. ### list_customers Retrieves a full list of all customers currently in your Prismatic environment. ### deploy_instance Pushes configuration changes live for a specific integration instance. ### fetch_logs Pulls execution logs for specific time ranges to help you find and fix integration errors. ### run_graphql Executes custom GraphQL queries to get specific data from the Prismatic schema. ## Prompt Examples **Prompt:** ``` List all customers who have an active integration. ``` **Response:** ``` I've pulled your active customer list. Here are the accounts currently running integrations: | Customer Name | ID | Status | | :--- | :--- | :--- | | **Acme Corp** | `cust_123` | Active | | **Globex** | `cust_456` | Active | | **Soylent Corp** | `cust_789` | Active | All three accounts are currently healthy and running on the latest instance versions. ``` **Prompt:** ``` Deploy instance inst_98765 and tell me when it's done. ``` **Response:** ``` I've triggered the deployment for instance `inst_98765`. **Deployment Status:** - **Target:** inst_98765 - **Action:** Configuration Update - **Status:** Success The configuration changes are now live in the runtime environment. ``` **Prompt:** ``` Fetch logs from 10:00 AM to 10:10 AM today for the Acme account. ``` **Response:** ``` I've retrieved the logs for the Acme account during that window. Here is the summary: - **Total Events:** 12 - **Successes:** 11 (Webhook Received, Data Sync) - **Failures:** 1 **Error Detail:** - **Type:** Execution Error - **Component:** Slack - **Message:** Connection timeout. The Slack component failed to send the notification. ``` ## Capabilities ### Create customer records Add new customer entries with custom external IDs for easier mapping. ### Deploy integration instances Push configuration changes live to a specific customer's setup. ### Pull execution logs Get timestamped logs to see exactly what happened in a workflow. ### Run custom GraphQL queries Access any part of the Prismatic schema, including actions and components. ### List all customers Get a full view of everyone currently using your integration platform. ### Set up new instances Generate unique integration environments for specific accounts. ## Use Cases ### Debugging a failed webhook A support agent asks the AI to pull logs from the last 10 minutes to find a timeout error in a specific workflow. ### Mass onboarding for enterprise A product manager asks the AI to create 50 new customer records and instances for a new large contract. ### Schema exploration for devs An engineer uses the agent to run a GraphQL query to see all available actions for a specific component. ### Rapid deployment check A developer asks the AI to deploy instance inst_98765 and confirm it is live in the runtime environment. ## Benefits - Stop hunting for customer IDs because you can use list_customers to see everything in one list. - Deploy changes faster by using deploy_instance to push live updates without manual dashboard navigation. - Debug issues in seconds by using fetch_logs to pull specific timestamped data for failed workflows. - Access deep data points instantly with run_graphql to query any part of your Prismatic schema. - Onboard new users faster by using create_customer and create_instance to set up accounts in one go. ## How It Works The bottom line is you get to manage your entire iPaaS backend using just a chat interface. 1. Connect your Prismatic API token to the Connector. 2. Give your AI agent a command like deploy the instance for Acme Corp. 3. The agent executes the request and confirms the deployment is live. ## Frequently Asked Questions **What can I do with the Prismatic MCP?** You can manage your entire Prismatic environment using natural language. This includes creating new customers, deploying integration instances, and fetching execution logs directly through your AI client. **How do I use Prismatic MCP to manage my customers?** You can simply ask your AI agent to list all customers or create a new one. It handles the API calls to add records and assign external IDs for you. **Can Prismatic MCP help me debug failed workflows?** Yes, it can pull specific execution logs based on timestamps. This helps you see exactly what happened during a failure without digging through a dashboard. **Does Prismatic MCP support GraphQL queries?** It does. You can ask your agent to run custom GraphQL queries to pull specific data from any part of your Prismatic schema, including actions and components. **How do I deploy instances using Prismatic MCP?** Just tell your agent which instance ID you want to deploy. It will trigger the deployment and push your configuration changes live immediately. **Can I use Prismatic MCP to see my integration logs?** Absolutely. You can request logs for specific time ranges to monitor your integration health or troubleshoot issues for your customers. **Can I execute custom GraphQL queries if a specific tool isn't available?** Yes! Use the `run_graphql` tool to execute any valid Prismatic GraphQL query or mutation. This gives you full access to the schema, including actions, components, and user data. **How do I make sure my integration changes are actually live?** After creating or updating an instance, you must use the `deploy_instance` tool with the specific Instance ID. This triggers the deployment process required for changes to take effect. **Is it possible to filter logs by a specific time range?** Yes, the `fetch_logs` tool requires `timestampGte` and `timestampLte` parameters in DateTime format, allowing you to pinpoint exactly when an event occurred.