# dbt Job Failure Debugging

> Test whether the LLM can effectively use MCP tools to diagnose a dbt Cloud job failure.

- Skill: `tools-only/dbt-job-failure-debugging` (Agent Skill, multi-file: 3 files)
- Install (CLI): `npx skillmds@latest add tools-only/dbt-job-failure-debugging`
- Raw SKILL.md: https://api.skillmd.com/api/skills/tools-only/dbt-job-failure-debugging/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Web & Frontend
- Author: tools-only (https://skillmd.com/u/tools-only)
- Updated: 2026-09-29
- Page: https://skillmd.com/skills/tools-only/dbt-job-failure-debugging

---

# dbt Job Failure Debugging

## Goal
Test whether the LLM can effectively use MCP tools to diagnose a dbt Cloud job failure.

## Expected Behavior
The LLM should:
1. Use the dbt MCP server to list recent job runs
2. Identify the failed run from January 14th
3. Get detailed error information using `get_job_run_error`
4. Classify the error type and investigate appropriately
5. Either find the root cause or document what was checked

## Evaluation Criteria
- **Tool Usage**: Did it use the MCP tools effectively?
- **Systematic**: Did it follow a logical debugging process?
- **Documentation**: Did it document findings appropriately?

## Prerequisites
This scenario requires:
- dbt MCP server running with valid credentials
- Access to a dbt Cloud account with job history