# Goldsky MCP for AI Agents AI Agent Connect > Goldsky (Web3 Data Indexing & Subgraphs) MCP lets you manage blockchain data pipelines and subgraphs directly through your AI agent. You can create, monitor, and update your indexing infrastructure using natural language instead of jumping between different dashboards or writing manual configuration files every time you need to tweak a pipeline. ## Overview - **Category:** developer-tools - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_vKwCn24rK9jiYRm7bJaactLmNUpPZSSNAnWrQ0l6/ai-agent-connect - **Tags:** web3, subgraphs, data-indexing, ethereum, real-time-data ## Description Managing blockchain data is usually a headache of switching between terminal windows and complex web consoles. If you're building a dApp, you need your data to be fresh, but keeping track of multiple indexing pipelines across different chains is a chore. This Connector changes that by letting you talk to your Goldsky infrastructure like it's a teammate. Instead of hunting for a specific pipeline ID to check its status or manually editing JSON files to deploy updates, you just ask your AI client to do it for you. You can see exactly which pipelines are running, which ones hit an error, and what the internal state of your data flow looks like in real time. It handles the heavy lifting of infrastructure management so you can stay focused on the actual code. It's one of the many ways Vinkius helps you get your tools talking to each other without the friction of traditional setups. You get a clear view of your indexing health and the ability to make instant changes to your data flow without leaving your workspace. ## Tools ### delete_pipeline Permanently removes a pipeline from your project. Use this to clean up old configurations. ### get_pipeline_error_count Shows how many errors occurred in a specific timeframe. It helps you spot spikes in indexing failures quickly. ### get_pipeline_logs Pulls the latest execution logs for a pipeline. You can use this to see exactly what happened during a specific block sync. ### get_pipeline_state Shows the internal state of a pipeline. This is useful for seeing where a process stalled. ### get_pipeline_status Tells you if a pipeline is running, paused, or failed. Use it for a quick health check. ### get_pipeline Fetches all the details for a specific pipeline. It's the go-to for inspecting your current configuration. ### list_pipelines Shows every pipeline currently in your project. Use it to get an overview of your entire indexing setup. ### pause_pipeline Stops a running pipeline immediately. Use this to save on costs or prevent data overlap during maintenance. ### restart_pipeline Reboots a pipeline that's stuck or needs a refresh. This is the quickest way to get a stalled sync moving again. ### resume_pipeline Starts a pipeline that was previously paused. It lets you pick up where you left off without a full restart. ### validate_pipeline Checks your JSON definition for errors before you deploy. This prevents broken configurations from breaking your live data. ### create_pipeline Deploys a brand new pipeline based on your requirements. You can create new data flows with just a few words. ## Prompt Examples **Prompt:** ``` List all my active Goldsky pipelines. ``` **Response:** ``` I've retrieved your pipelines. You have 3 active: 'eth-mainnet-swaps', 'polygon-nft-events', and 'base-bridge-monitor'. Which one would you like to inspect? ``` **Prompt:** ``` Check the status and error count for the pipeline 'eth-mainnet-swaps'. ``` **Response:** ``` The pipeline 'eth-mainnet-swaps' is currently **RUNNING**. I detected 0 errors in the last time window. Everything looks healthy. ``` **Prompt:** ``` Show me the recent logs for the 'polygon-nft-events' pipeline. ``` **Response:** ``` Fetching logs for 'polygon-nft-events'... * [INFO] Successfully indexed block 5420123 * [INFO] Processed 12 Transfer events * [DEBUG] Sink connection stable. ``` ## Capabilities ### Create new pipelines Deploy new Web3 data flows using your existing JSON definitions. ### Check pipeline status See if your data flows are running, paused, or failed in one step. ### Fetch execution logs Pull recent logs to see exactly what happened during a specific block sync. ### Pause or resume flows Stop or start your indexing pipelines to manage resource usage instantly. ### Validate configurations Check your JSON definitions for errors before you push them live. ### Monitor error counts Get a count of errors within a specific timeframe to spot spikes quickly. ### Remove old pipelines Delete obsolete pipelines from your project to keep your workspace clean. ## Use Cases ### Debugging a stalled sync A developer notices a subgraph isn't updating. They ask the agent to check the logs for 'polygon-nft-events' and see why it stopped. ### Cost management An engineer wants to turn off a test pipeline over the weekend. They tell the agent to pause the 'test-net-monitor' to save on Goldsky credits. ### Rapid deployment A team needs a new indexing flow for a new token. They describe the requirements to the agent, which then creates the pipeline using the JSON definition. ### Health auditing A protocol lead wants to know if the main bridge indexer is healthy. They ask the agent to check the error count for the last hour. ## Benefits - Stop manual JSON editing by using create_pipeline to deploy new Web3 data flows instantly. - Reduce downtime by using restart_pipeline to quickly fix stalled blockchain indexing processes. - Save on infrastructure costs by using pause_pipeline to stop data flows when they aren't needed. - Catch configuration bugs early by using validate_pipeline before you push changes to production. - Debug faster by pulling real-time logs with get_pipeline_logs instead of hunting through a web console. - Monitor health at a glance using get_pipeline_status to see if your data is flowing correctly. ## How It Works The bottom line is you get full control over your Web3 data infrastructure through a simple chat interface. 1. Connect your Goldsky API Key to the Connector. 2. Describe the pipeline action you want to take in your AI client. 3. Your agent executes the command and gives you the status or logs immediately. ## Frequently Asked Questions **Can I use the Goldsky MCP to create new subgraphs?** Yes, you can create and deploy new pipelines using natural language. Just describe the requirements to your agent and it will handle the deployment for you. **How does the Goldsky MCP help with debugging?** It lets you pull execution logs and error counts directly into your chat. You can see exactly what went wrong during a specific sync without leaving your workspace. **Can I pause my data indexing to save money?** Yes, you can use the pause and resume tools to stop data flows when they aren't needed, helping you manage your infrastructure costs more effectively. **Does the Goldsky MCP work with Cursor and VS Code?** Yes, this Connector is compatible with any MCP-enabled client, including Claude, Cursor, and Windsurf. **How do I see if my pipelines are failing?** You can ask your agent to check the status or the error count for any specific pipeline. It will report if the flow is running, paused, or failed. **Can I check my pipeline logs without a dashboard?** Absolutely. Your agent can fetch and display the latest logs for any pipeline, allowing you to troubleshoot issues directly in your chat interface. **Can I check if my pipeline configuration is valid before deploying it?** Yes! Use the `validate_pipeline` tool with your definition object. It will check your sources, transforms, and sinks for errors without actually starting a new pipeline. **How do I monitor errors in a running pipeline?** You can use `get_pipeline_error_count` to see the number of issues in a time window, or `get_pipeline_logs` to fetch the actual execution logs for detailed debugging. **Is it possible to temporarily stop a pipeline without deleting it?** Absolutely. Use the `pause_pipeline` tool to stop execution. When you are ready to start again, simply use the `resume_pipeline` tool.