incident-triage

Diagnose RAG pipeline failures by locating the failing stage and root cause.

5|1|Updated May 9, 2026
One-click install
npx skills add https://github.com/ql-link/LinkRag --skill incident-triage-ql-link
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: incident-triage
Source: https://github.com/ql-link/LinkRag/tree/main/.ai/skills/incident-triage
Command: npx skills add https://github.com/ql-link/LinkRag --skill incident-triage-ql-link

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill standardizes fault triage for the RAG parsing and recall pipeline, enabling engineers to quickly locate the failing stage and determine root causes.

Core Features & Use Cases

  • Structured fault localization: map a failure to the pipeline stage (parse_task → six-stage pipeline) and classify root cause as configuration drift, data inconsistency, code defect, or external dependency.
  • Actionable remediation: provide concrete fix steps and, when needed, data reset/recovery actions to restore pipeline health.
  • Use Case: When logs show parse_task failures, MQ consumption stalls, or vector/search errors, run triage to identify the failing segment and recommended recovery actions.

Quick Start

Provide a step-by-step triage guide by reviewing parse_task_consumer, pipeline stages, and config files to identify the failing segment and recommended fixes.

Frequently Asked Questions about incident-triage

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I triage RAG pipeline failures when parse_task logs show errors?

To triage RAG pipeline failures, map the parse_task errors to the specific six-stage pipeline segment and classify the root cause as configuration drift, data inconsistency, code defect, or external dependency to determine the exact failing stage.

Why does MQ consumption stall in my RAG recall pipeline?

MQ consumption stalls in the RAG pipeline occur due to configuration drift, data inconsistency, code defects, or external dependency failures. Running a standardized fault triage on the consumer logs and config files isolates the exact stage and determines the necessary recovery action.

What is the best way to resolve Qdrant or Elasticsearch errors during RAG recall?

The best way to resolve Qdrant or Elasticsearch errors during RAG recall is to run a standardized fault triage that locates the failing segment and provides actionable fixes including configuration adjustments, code corrections, or data resets with safety checks.

Can I get actionable fix steps for data inconsistency issues in a RAG pipeline?

Yes, you can get actionable fix steps for data inconsistency issues in a RAG pipeline. The triage process identifies the root cause and provides concrete remediation actions, including data reset and recovery procedures with repeatable steps to restore pipeline health.

Does RAG fault triage work for both configuration drift and external dependency outages?

Yes, RAG fault triage works for both configuration drift and external dependency outages. It classifies the root cause across the parsing and recall pipeline stages and provides targeted recovery actions to address the specific failure type.

How do I locate the exact failing stage when a RAG pipeline is stalled?

To locate the exact failing stage in a stalled RAG pipeline, review the parse_task consumer, the six-stage pipeline flow, and configuration files. This step-by-step triage identifies the failing segment and recommends the appropriate fixes.