# Sumo Logic MCP for AI Agents AI Agent Connect > Sumo Logic connects your AI agent directly to your enterprise log, metrics, and collector infrastructure. Instead of navigating dashboards, you use natural conversation to run deep search jobs, monitor system health in real time, and manage all data sources using plain English commands. ## Overview - **Category:** cloud-infrastructure - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_n0vvKHAB1OwS9Ns2pKbhWu0crSxY8a6AVPotcafn/ai-agent-connect - **Tags:** logs, metrics, siem, observability, devops ## Description This Connector lets your AI client treat your observability platform like a command line interface (CLI). You can talk to it about your logs the way you'd talk to an experienced DevOps engineer. Need to find why checkout latency spiked last Tuesday? Just ask. It handles initiating complex, asynchronous search jobs and gives you raw log messages when they finish. It’s not just for reading logs; you manage the infrastructure itself. You can use it to list all your collectors or even create a new hosted collector if one's offline. Want better visibility? You configure data sources right through your agent. The whole process, which used to involve jumping between admin tabs and query builders, now happens in a single chat session. Because this Connector is available on Vinkius, you connect once from any compatible client and gain access to all these critical infrastructure tools. ## Tools ### create_source Defines and adds a new specific data source within an existing collector. ### create_user Adds a brand new user account to the organization for access control. ### delete_collector Removes a specified data collection endpoint from your environment. ### delete_search_job Cancels and removes an ongoing or completed search job to free up resources. ### delete_source Removes a defined data source from its associated collector. ### delete_user Deactivates and removes an existing user account. ### execute_metrics_query Runs a query against time-series data to gather specific performance metrics. ### get_search_job_messages Pulls the raw, unaggregated text messages from a search job once it completes. ### get_search_job_records Retrieves structured and aggregated records after a search job has finished processing. ### get_search_job_status Checks if an initiated log search job is still running or if it's complete. ### get_source Retrieves the specific details for a defined data source within a collector. ### list_collectors Lists every installed and hosted collector currently configured in your account. ### list_users Generates a list of every user account currently configured in the organization. ### update_collector Modifies the settings or configuration details of an existing collector. ### update_source Changes the properties or definition of a specific data source. ### get_collector Retrieves all the detailed information for a single collector using its unique ID. ### create_hosted_collector Sets up a new collector hosted within the Sumo Logic environment. ### create_search_job Starts an asynchronous, deep search job for logs that will require time to process. ### list_sources Shows all defined data sources available within a specific collector ID. ## Prompt Examples **Prompt:** ``` List all my Sumo Logic collectors. ``` **Response:** ``` I've retrieved your collectors. You have 3 active collectors: 'Prod-Receiver' (ID: 1023), 'Staging-Logs' (ID: 1024), and 'Cloud-Hosted' (ID: 1025). ``` **Prompt:** ``` Search for 'error' in logs for the last 15 minutes. ``` **Response:** ``` I've initiated a search job (ID: job_98765). I will poll the status and let you know when the results are ready. ``` **Prompt:** ``` Get the records for search job ID job_98765. ``` **Response:** ``` The search job is complete. I found 12 records matching 'error'. Most common message: 'Connection timeout at 10.0.0.1'. ``` ## Capabilities ### Run complex log searches Initiate deep, asynchronous search jobs across your entire data set and retrieve the resulting raw logs or aggregated records. ### Monitor system metrics Execute time-series queries to check performance health, such as CPU utilization or request counts, in real time. ### Manage data collection infrastructure List, create, update, or delete collectors and manage the specific sources feeding into them. ### Administer user accounts Maintain security by listing or creating new users within your Sumo Logic organization. ## Use Cases ### Debugging a production outage A DevOps engineer gets paged at 3 AM. They ask their agent, 'Show me all connection errors from the last two hours.' The agent automatically initiates a search job and later provides the raw logs via `get_search_job_messages`, pinpointing the failing service immediately. ### Onboarding a new data pipeline An SRE needs to monitor traffic from a new microservice. They ask their agent to list collectors, identify where it should go, and then use `create_hosted_collector` followed by `create_source` to get the required log flow running. ### Auditing user access A Security Analyst suspects unauthorized activity. They ask their agent to list all users and check who has administrator rights, using `list_users` to scope the investigation immediately. ### Capacity planning for growth The team realizes they need more monitoring capacity. They use the Connector to monitor current system metrics (`execute_metrics_query`) and then request status updates on existing collectors using `get_search_job_status` to ensure everything is healthy before scaling. ## Benefits - Instantly search deep log data. Instead of running a job and then having to poll its status manually, your agent manages the entire lifecycle—from using `create_search_job` to pulling results with `get_search_job_records`. - Manage infrastructure without switching tabs. You can use `list_collectors` or `update_collector` directly through natural conversation, handling data ingestion setup faster than ever before. - Pinpoint performance bottlenecks using metrics. Execute precise queries via `execute_metrics_query` to monitor system health and track resource usage in real time. - Maintain compliance and security easily. Use the agent to list all users (`list_users`) or create new accounts (`create_user`), ensuring your access controls are always up-to-date. - Full lifecycle control over data sources. Need to adjust what's being monitored? You can use `get_source` to check details, and then `update_source` to fix the configuration. ## How It Works The bottom line is that it turns complex, multi-step platform administration into simple conversation. 1. Subscribe to this Connector and enter your required Sumo Logic Access ID, Access Key, and API URL. 2. Your AI client establishes the connection. It authenticates and verifies access to your log and metrics environment. 3. You ask a question—for example, 'What were the error logs in the last hour?' Your agent executes the necessary search job and presents you with the results. ## Frequently Asked Questions **How does the Sumo Logic MCP help me debug production issues?** It lets you run complex searches and retrieve raw log messages conversationally. Instead of running a job, waiting hours, and then fetching results manually, your agent manages the whole process, giving you instant visibility into what broke. **Can I use this Connector to manage my data collectors?** Yes. You can list all collectors with simple commands, create brand new hosted ones using `create_hosted_collector`, or delete old endpoints if they're no longer needed. **Is the Sumo Logic MCP good for monitoring performance metrics?** It is. You can execute precise time-series queries to monitor key system health indicators, like CPU usage or request count. This gives you quantitative data on your infrastructure's real capacity. **Do I need to be a developer to use the Sumo Logic MCP?** No. The whole point is that you don't have to know API calls. You talk naturally, and the agent handles translating your request into the necessary technical action. **What if I need to add a new data source? Can the Sumo Logic MCP do it?** Absolutely. You can use the Connector to list available sources and then define or update them, ensuring that all your required logs are flowing into the system properly. **How do I start a new log search and get the results?** First, use `create_search_job` to start the search. Then, poll `get_search_job_status` until it's complete. Finally, use `get_search_job_messages` or `get_search_job_records` to retrieve the data. **Can I see all my configured data collectors?** Yes, use the `list_collectors` tool to retrieve a list of all Installed and Hosted Collectors in your account, including their IDs and status. **How do I monitor system performance metrics?** You can use the `execute_metrics_query` tool to run specific metrics queries and get real-time data on your infrastructure's health.