# Temporal MCP for AI Agents AI Agent Connect

> Temporal MCP lets you monitor and manage distributed workflows in Temporal Cloud natively via your AI agent. It allows you to inspect active runs, pull execution histories for debugging, and query complex visibility logs using natural language. Instead of hunting through the Temporal UI, you can just ask your agent to find a specific failed workflow or check a namespace's retention period.

## Overview
- **Category:** ship-it
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_EnU6gjOxc9KSHxcvKKIpGvDNCsHHJOBTVxTch47l/ai-agent-connect
- **Tags:** workflow-orchestration, distributed-systems, durable-execution, event-history, debugging, cloud-automation

## Description

When you're managing distributed systems, things go wrong in ways that are hard to pin down. Usually, this means jumping between browser tabs, digging through logs, and trying to piece together why a specific workflow failed or where a timer got stuck. This Connector changes that by letting you talk directly to your Temporal cluster from your favorite editor or chat window. You can ask your agent to find every running workflow in a specific namespace or grab the full event history for a job that crashed at 3 AM. It's about getting the right data without the manual overhead of navigating a complex dashboard. Whether you're auditing permissions or troubleshooting a production incident, you can just describe what you're looking for and get the facts immediately. It fits right into your existing workspace, and since it's hosted on the Vinkius catalog, getting it connected to your tools takes just a few minutes.

## Tools

### get_workflow_history
Pull the full sequence of events and activities for a specific workflow execution to find exactly where it failed.

### get_namespace_details
Check your current namespace's configurations, metadata, and retention periods in one go.

### get_workflow_details
Get the specific status and metadata for a single workflow run to see its current state.

### list_schedules
See all recurring workflows and their next scheduled execution times across your cluster.

### list_search_attributes
View the custom search attributes available in your namespace to understand your querying options.

### list_workflows
Get a high-level list of all workflow IDs, run IDs, and statuses in your namespace.

### search_workflows
Run complex visibility queries to find specific runs based on status, type, or other attributes.

## Prompt Examples

**Prompt:** 
```
Why did the last run of the 'OrderProcessing' workflow fail?
```

**Response:** 
```
I checked the history for the most recent `OrderProcessing` run and found the following:

**Failure Details:**
- **Activity:** `AuthorizePayment`
- **Error:** `InsufficientFundsException`
- **Timestamp:** 2026-07-15 14:22:01 UTC

The workflow stopped because the payment gateway returned a 402 error. Would you like me to pull the full history of the preceding activities?
```

**Prompt:** 
```
List all active schedules and tell me when the database backup is due.
```

**Response:** 
```
You have 2 active schedules:

| Schedule Name | Frequency | Next Execution |
| :--- | :--- | :--- |
| `WeeklyReport` | Every Monday 9AM | 2026-07-20 09:00:00 |
| `DB_Backup` | Daily Midnight | 2026-07-16 00:00:00 |

The next database backup is scheduled to start in approximately 4 hours.
```

**Prompt:** 
```
Show me all running workflows in the production namespace.
```

**Response:** 
```
I found 4 active workflows in your current namespace:

- `UserSignup_7721` (Running)
- `InvoiceGen_1102` (Running)
- `DataSync_9982` (Running)
- `EmailBlast_4431` (Running)

Do you want me to get the specific details for any of these runs?
```

## Capabilities

### Inspect active and completed workflow statuses
See a list of all workflow IDs and their current statuses in your namespace.

### Pull full event histories for deep debugging
Retrieve the exact sequence of events and activities for a specific run.

### Execute complex visibility queries
Search for specific workflows using natural language instead of manual filtering.

### View namespace configurations
Check your current namespace's metadata and retention periods instantly.

### Browse recurring schedules
See all active schedules and their predicted next execution times.

## Use Cases

### Failed Job Debugging
An engineer asks, "Why did the InvoiceWorkflow fail?" The agent uses get_workflow_history to show the exact activity that threw an error.

### Production Incident Response
An SRE asks for all running workflows in a specific namespace. The agent uses list_workflows to provide an immediate status report.

### Schedule Audit
A platform team wants to know when the next backup runs. The agent uses list_schedules to show the next execution times.

### Namespace Verification
A dev needs to check if a namespace has the correct retention. The agent uses get_namespace_details to pull the config.

## Benefits

- Stop manual searching by using search_workflows to find specific failures instantly.
- Debug faster by pulling full execution logs with get_workflow_history instead of hunting through the UI.
- Audit your environment easily by checking namespace metadata with get_namespace_details.
- Stay on top of recurring jobs by viewing all upcoming runs with list_schedules.
- Reduce context switching by querying your cluster directly from Cursor or Claude.
- Identify bottlenecks quickly by running complex visibility queries using natural language.

## How It Works

The bottom line is you get instant, conversational access to your Temporal cluster data without leaving your workspace.

1. Subscribe to the Temporal MCP through the Vinkius platform.
2. Enter your Temporal Address, Namespace, and API Key.
3. Ask your agent questions about your workflows or query cluster states.

## Frequently Asked Questions

**Can I use the Temporal MCP to debug failed workflows in Claude or Cursor?**
Yes. This Connector allows your AI client to query your Temporal cluster directly. You can ask it to find specific failed runs or pull their full history without leaving your editor.

**Does the Temporal MCP work with self-hosted clusters?**
Yes, it works with both Temporal Cloud and self-hosted clusters. You just need to provide your Temporal Address and the correct API credentials.

**How does the Temporal MCP help with production incidents?**
It speeds up incident response by letting you query visibility logs and list active workflows using natural language. This eliminates the need to navigate complex UIs during a crisis.

**Can the Temporal MCP check my namespace configurations?**
Yes. You can ask your agent to retrieve namespace details, such as retention periods and metadata, to ensure your environment is configured correctly.

**Will the Temporal MCP show me my recurring cron jobs?**
Yes. It can list all workflow schedules and even tell you the predicted next execution time for your recurring tasks.

**Do I need to write any code to use the Temporal MCP?**
No code is required. Once you subscribe and provide your credentials, you can interact with your Temporal data through plain English conversation with your AI agent.

**Can my AI agent debug a failed Temporal workflow run directly?**
Yes. Instead of browsing the Temporal UI to manually dissect the execution, feed the workflow ID directly to your AI agent. It can seamlessly fetch the entire event history, identify the exact panic line or activity timeout, and immediately suggest code fixes within your IDE.

**How complex can the visibility searches be via natural language?**
Very complex. Because your agent understands your codebase, you can ask 'Find all workflows of type ProcessPayment that failed in the last 2 hours'. The agent formulates the perfect SQL-like syntax for the backend, parses the matching pages, and surfaces exactly which transactions failed.

**Can it read scheduled jobs and Cron jobs configurations inside Temporal?**
Absolutely. You can request a list of all active schedules mapped to your namespace. Your agent unpacks JSON configurations, showing definitions of workflow recurrence, enabling you to detect missing crons without interrupting your workflow.