--- title: "Triage a data-quality failure with full context - Atlan MCP Cookbook" url: "https://atlan.com/mcp-cookbook/triage-a-data-quality-failure-with-full-context/" description: "Enrich quality failures with lineage, ownership, and downstream impact for faster triage." --- > Atlan is hosting Context Conference, bringing together the leaders and builders at the frontier of giving AI the context it needs to understand their business. It runs online on October 28, 2026, from 11:00 AM to 2:00 PM ET. Atlan co-founder Prukalpa Sankar opens and closes the day. Leaders from AstraZeneca, BNY and Verizon share why they invest in context and what they get from it. Registrants get early access to The AI Context Gap, a new study from MIT Technology Review Insights. Register: https://atlan.com/context-conference/ Recipe 8 in [The Atlan MCP Cookbook](https://atlan.com/mcp-cookbook/). Enrich quality failures with lineage, ownership, and downstream impact for faster triage. Use case: Change & fix data safely / Action: Make changes / Persona: Data Engineer / Effort: Moderate / Runs in: any AI tool Requires Atlan MCP connected to your AI tool: [set up Atlan MCP](https://docs.atlan.com/product/capabilities/atlan-ai/how-tos/remote-mcp-overview). ## 1. The task **Scenario.** orders_fact failed a null check on order_id this morning. The raw alert says something broke; it does not say who owns the table or which dashboards break next. Whoever is on call needs that context now, not after twenty minutes of digging. The ask: - The owners of orders_fact and its DQ rule failures ([Data Quality Studio](https://docs.atlan.com/product/capabilities/governance/data-quality)) - The downstream dashboards it feeds - A warning announcement on the table with the failure details Hero prompt: "The orders_fact table failed a null check on order_id this morning. Find its owners and downstream dashboards, then add a warning announcement with the failure details so the team can triage." Step by step: 1. Pull the failing asset's owners, tags, announcements, and its current DQ rule failures and incidents. Prompt: "The [table] failed a [null] check on [column] this morning - show its owners, tags, announcements and its current DQ rule failures and incidents." 2. Trace its downstream consumers. Prompt: "Trace the downstream dashboards and tables that depend on [table], so I know what's at risk." 3. Post a warning announcement with the failure details. Prompt: "Add a warning announcement to [table] with the failure details so the team can triage." ## 2. What you'll give it Required context: - The failing asset and what failed (orders_fact, null check on order_id) - Lineage populated, so downstream consumers are real - Permission to post announcements Optional context: - The check's run details or a link, to paste into the announcement - The on-call channel, if you also want a ping and not just the on-asset notice - Data Quality Studio (or another data quality connector) enabled on the failing asset, to pull its actual DQ rule failures, not just a posted status ## 3. What Atlan creates Outputs: - The owners of orders_fact, its DQ rule failures from Data Quality Studio, and the downstream dashboards at risk - A warning announcement on orders_fact carrying the failure details - A triage-ready picture: what broke, who owns it, what it hits **Summary.** The alert arrives wrapped in the context that makes it actionable. On-call sees blast radius and ownership in one place, on the asset itself, instead of reconstructing it under pressure. **Writes to Atlan.** Changed: one warning announcement on the failing table. Left untouched: the data, the check, and every downstream asset, which are surfaced but not edited. ## 4. Follow-up prompts - **Warn the dependents:** "Post a short heads-up announcement on each downstream dashboard that this table failed a check today." - **Draft the update:** "Write a two-line status update for the on-call channel summarizing the failure and who is affected." - **Find the pattern:** "Show the current quality rules and any open incidents on this table, so I can see if it's a repeat offender." ## 5. Tips & troubleshooting **Cook's note.** Post the announcement on the failing asset itself. Everyone downstream sees it in context. - Match the tool to the work. During an incident, run it in whatever AI tool is already open; the value is speed, and the announcement is a small, safe write. - Automate or chain it. Chain it to your quality tool: on any check failure, fire this recipe so the announcement is posted before a human even looks. - Ask what it changed. Add "confirm which asset you posted the announcement on." One line, so you know exactly where the notice landed. ## Related recipes - [Warn downstream owners before a change lands](https://atlan.com/mcp-cookbook/warn-downstream-owners-before-a-change-lands/): Find every downstream consumer of a changing asset and warn their owners first. - [Trace a broken dashboard to its root cause](https://atlan.com/mcp-cookbook/trace-a-broken-dashboard-to-its-root-cause/): Trace lineage upstream from a broken dashboard to find the source of wrong numbers. - [Keep asset metadata fresh](https://atlan.com/mcp-cookbook/keep-asset-metadata-fresh/): Catch stale certifications, missing owners, and outdated descriptions early. [Browse all recipes](https://atlan.com/mcp-cookbook/) / [Set up Atlan MCP](https://docs.atlan.com/product/capabilities/atlan-ai/how-tos/remote-mcp-overview)