# Zeplo MCP for AI Agents AI Agent Connect > Zeplo (Queue & Background Job API) lets you manage your background jobs, webhooks, and scheduled tasks directly through your AI agent. Instead of jumping between the Zeplo dashboard and your code, you can ask your agent to enqueue requests, check why a webhook failed, or schedule new recurring tasks. It turns your background job infrastructure into a conversational interface for faster debugging and deployment. ## Overview - **Category:** developer-tools - **Price:** Free - **Endpoint:** https://edge.vinkius.com/undefined/ai-agent-connect - **Tags:** queue-management, background-jobs, webhooks, scheduling, serverless ## Description This Zeplo MCP lets you manage your background jobs, webhooks, and scheduled tasks directly through your AI agent. Managing these tasks usually feels like a chore of constant context switching. You have to leave your IDE, open a dashboard, find the right queue, and hunt for a specific job ID just to see why a webhook failed or to trigger a retry. This connection changes that by letting you handle your Zeplo infrastructure using plain English. You can tell your AI agent to set up a new recurring schedule, enqueue a batch of test requests, or pull up the logs for a failing job without ever opening a browser tab. It makes your background processing pipeline feel like an extension of your conversation with your agent. When you're using the Vinkius marketplace to connect your tools, this integration ensures that your AI isn't just a coding partner but a real operator for your production environment. It handles the heavy lifting of monitoring and lifecycle management so you can stay focused on building features rather than babysitting queues. Instead of manually checking the status of every single request, you get a clear view of your queue health and execution history. You can quickly identify bottlenecks, manage retries for failed jobs, and ensure your production webhooks are firing as expected. This gives you a way to oversee your entire background task flow without the friction of multiple tabs and manual data entry. ## Tools ### create_queue Create a new queue for your background tasks. This helps you organize different types of jobs into specific pipelines. ### delete_schedule Remove an existing schedule from your workspace. Use this when you no longer need a recurring task to run. ### get_queue See the specific metrics and details for a single queue. This helps you monitor the health and volume of your tasks. ### cancel_request Stop a request that is currently pending or scheduled. Use this to prevent a job from running if it is no longer needed. ### create_schedule Set up a new recurring or delayed task schedule. You can define when jobs should run without writing any cron logic. ### create_token Generate a new workspace token for secure access. This is the first step to giving your agent permission to manage your data. ### delete_queue Remove a queue from your workspace entirely. This helps you clean up old or unused processing pipelines. ### enqueue_request Send a new HTTP request to be processed by the queue. This is how you add new work for your background processes to handle. ### get_request Look up the timeline and details for a specific request. It shows you exactly what happened during a job execution. ### invite_team_member Add a new person to your workspace team. This lets you manage your workspace permissions through natural language. ### list_queue_logs View a paginated history of requests for a specific queue. It is the fastest way to see your recent success and failure rates. ### list_queues See all queues currently active in your workspace. Use this to get a high level view of your entire background infrastructure. ### list_schedules See all active schedules in your workspace. This helps you keep track of all your recurring automated tasks. ### list_team View the members currently in your workspace team. Use this to check who has access to your workspace. ### list_tokens View all tokens currently active in your workspace. This helps you manage your security credentials in one place. ### update_queue Modify the settings of an existing queue. You can change how a queue behaves without deleting and recreating it. ### update_schedule Pause, resume, or edit an existing schedule. This gives you quick control over when your automated tasks run. ## Prompt Examples **Prompt:** ``` Show me the status of the last 3 jobs in the 'email_sync' queue. ``` **Response:** ``` I've checked the `email_sync` queue for you. Here are the most recent jobs: | Job ID | Status | Timestamp | Payload | | :--- | :--- | :--- | :--- | | `zeplo_123` | **Success** | 2 mins ago | `{"user_id": 44}` | | `zeplo_124` | **Pending** | Just now | `{"user_id": 45}` | | `zeplo_125` | **Failed** | 5 mins ago | `{"user_id": 46}` | It looks like the last job failed. Would you like me to trigger a retry for `zeplo_125`? ``` **Prompt:** ``` I need to schedule a GET request to my health check endpoint every 10 minutes. ``` **Response:** ``` I've created a new schedule for your health check endpoint. **Schedule Details:** * **Endpoint:** `https://api.service.com/health` * **Frequency:** Every 10 minutes * **Status:** Active I've set this up to trigger automatically. You can ask me to list all active schedules anytime to verify it. ``` **Prompt:** ``` Why did the last request to the payment gateway fail? ``` **Response:** ``` I retrieved the details for the most recent request to the payment gateway (`zeplo_998`). **Error Details:** * **Status Code:** 500 Internal Server Error * **Response Body:** `{"error": "Gateway timeout", "code": "GATE_01"}` * **Timeline:** The request reached the gateway but didn't receive a response within the 30s limit. Would you like me to retry this request now? ``` ## Capabilities ### Enqueue HTTP requests with retries Send new requests to your background queue with automatic retry logic. ### Create recurring CRON schedules Set up automated tasks that run on a recurring schedule. ### Fetch specific job details and payloads Retrieve the exact data and execution history for any specific job. ### Filter and list recent queue logs Browse through your recent job history to find successes and failures. ### Cancel pending or scheduled requests Stop a job from running before it starts or cancel a scheduled task. ### Create and manage workspace tokens Generate and manage security tokens for your workspace access. ### Invite team members to the workspace Add new people to your workspace team using natural language commands. ## Use Cases ### Debugging a Webhook A developer asks the agent to list the last 5 failed jobs to see why a payment notification didn't process. ### Production Recovery An engineer tells the agent to retry a specific failed job that timed out during a high-traffic period. ### Automated Reporting A product manager asks the agent to create a schedule that triggers a daily summary report every morning at 8 AM. ### Testing New Endpoints A developer enqueues a series of POST requests with different payloads to test a new API endpoint's reliability. ## Benefits - Stop hunting for job IDs by using `get_request` to see exactly what happened to a specific task. - Fix production issues faster by using `cancel_request` to stop problematic jobs instantly. - Automate your recurring tasks without code changes by using `create_schedule` to set up CRON jobs. - Monitor your infrastructure health in real-time using `list_queue_logs` to see success and failure rates. - Manage your team's access easily by using `invite_team_member` and `create_token` directly through your agent. - Organize your work pipelines by using `create_queue` and `update_queue` to manage different task types. ## How It Works The bottom line is you get to manage your entire background job lifecycle through a single chat interface. 1. Subscribe to the Zeplo MCP and provide your API key. 2. Tell your AI agent to perform an action like "enqueue a request" or "list failed jobs." 3. Receive immediate confirmation of the job status or the specific data you requested. ## Frequently Asked Questions **Can the Zeplo MCP help me debug failed webhooks?** Yes, it lets you see the logs and errors directly in your chat. You can ask your agent to pull up the history for a specific queue to see exactly why a webhook didn't process. **How do I schedule recurring tasks with Zeplo?** You can just tell your agent what you want to do. For example, ask it to run a GET request every hour, and it will use the scheduling tools to set that up for you automatically. **Can I cancel a job that's already in the queue?** Yes, you can stop any pending or scheduled requests. Just tell your agent which job ID or request you want to stop, and it will handle the cancellation for you. **Does this Connector support retries for failed tasks?** Yes, your agent can trigger retries for failed jobs manually. This is great for fixing production issues quickly without having to edit your application code. **How do I add people to my Zeplo workspace?** You can manage your workspace permissions through natural language. Just ask your agent to invite a new team member by name or email. **Can I see the history of my queue logs?** Yes, the agent can pull up a paginated history of all recent requests for any specific queue. This helps you monitor success and failure rates at a glance. **Can I schedule a job to run at a specific time in the future?** Yes! Use the `enqueue_job` tool and specify a delay or a schedule parameter. You can tell the AI 'Schedule this request for tomorrow at 10am' and it will handle the timestamp conversion for Zeplo. **How do I check why a specific background job failed?** You can use the `get_job` tool with the Job ID. The AI will retrieve the full execution log, including the response body and status code from your endpoint, helping you debug the failure instantly. **Is it possible to cancel a job that hasn't run yet?** Absolutely. Use the `cancel_job` tool with the target Job ID. This is useful for stopping scheduled tasks or queued jobs that are no longer necessary.