AI Agent Use Cases
Views —

AI Agent Use Cases

Added today

44All 189

Prompt to Prepare a Shopping Cart from a Recipe Photo

The author reports that Grok Bot receives a recipe photo, organizes the ingredients and quantities, and creates a shopping list by excluding ingredients already available at home.

Grok Bot
Life & Hobbies

1–20 of 189

DevWalkthrough

Agent Investigates Build Failures Using GitLab Logs and Monitoring Data

This is an example where an agent investigates build failures by checking GitLab logs and Grafana status to identify possible causes.

n8n

How it works

  1. Connect GitLab and Grafana MCP, as well as an HTTP query tool, and specify the failed job.
  2. The agent selects GitLab, Grafana MCP, and the HTTP tool to investigate the logs and status of the failed job.

Results

In the public example, the intended outcome is evidence for the build failure and a draft of a solution.

Notes

This is an example of investigating the cause of a failed build and drafting a solution. It does not include verifying tools that perform deployment recovery. The author reports checking public settings and tool connections, but did not execute them directly.

Source

n8n 공개 워크플로 #14129 Original post ↗

Example prompt · n8n

Goal: Diagnose a GitLab build failure using logs and monitoring data.

Inputs/Environment: [Failed job URL], [GitLab connection], [Grafana connection], [Scope of diagnosis].

Reference original: https://n8n.io/workflows/14129

Please prepare the settings and connections so that a test environment can be created using the public configuration from the original. If there are missing sub-workflows, APIs, or account permissions, mark them as incomplete until they are actually connected.

Connections and data to prepare: Connect GitLab, Grafana MCP, and an HTTP query tool, and specify the failed job.

Agent actions to implement: The agent selects GitLab, Grafana MCP, and the HTTP tool to investigate the logs and status of the failed job.

Scope of actions: This is an example of investigating the cause of a failed build and drafting a solution. It does not include verifying tools that perform deployment recovery.

The deliverable is evidence for the build failure and a draft solution. Please also organize the necessary setup steps and actual execution methods. After checking the results returned by the tools, select the next task, and separately note any data that could not be accessed or tasks that could not be executed.

Verification: Check that logs from other runs are not mixed in, and that confirmed causes are separated from still-suspected causes. Show the inputs used for verification, the actual outputs, and clearly distinguish items that have not yet been confirmed.