A spring clean for your test repository with Qase MCP

A QA team used Qase MCP, Claude Code, and a rule-based AI workflow to audit 1,209 manual test cases that had accumulated over years, reducing an estimated month of manual work for three engineers to about one week for a single engineer. The agent examined case steps and expected results, traced relevant product code, searched tests across E2E, API, integration, unit, frontend, and component layers, and recorded evidence-based coverage verdicts in a custom field. The audit identified cases already covered by automation, including backend tests not reflected in E2E coverage, found 27 duplicate cases by comparing their actual scenarios rather than titles, and marked seven cases as deprecated only after confirming their features no longer existed in the codebase. To limit AI errors, especially false claims of coverage, the team required named assertions, verified every linked test, restricted editable fields, trialed the first write in each batch, prohibited deletion, and used a low-confidence outcome when evidence was insufficient. The approach emphasizes that AI can efficiently perform repetitive repository analysis, while people retain responsibility for irreversible decisions such as deleting duplicates or archiving cases.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.