sumo-qa-deciding-approach

Classify QA change intents and route to the matching sub-skill.

5|1|Updated May 12, 2026
One-click install
npx skills add https://github.com/sumithr/sumo-qa --skill sumo-qa-deciding-approach
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sumo-qa-deciding-approach
Source: https://github.com/sumithr/sumo-qa/tree/main/skills/sumo-qa-deciding-approach
Command: npx skills add https://github.com/sumithr/sumo-qa --skill sumo-qa-deciding-approach

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Prevents wasted effort by selecting the correct QA approach for a change intent before scaffolding tests, writing plans, or proposing merges.

Core Features & Use Cases

  • First-step router: Loads only the classification and approach catalogues, then routes to the single matching sub-skill.
  • Shape + reachability guardrails: Determines change “shape” (single-change vs repo-wide vs docs/config/no-tests) and runs an orphan/deletion reachability gate before choosing any test-writing approach.
  • Strict routing discipline: Returns an internal routing payload with classification, approach, rationale, and next_action.skill, including STOP cases with no sub-skill handoff.

Quick Start

Use sumo-qa-deciding-approach at the start of any QA intent to route the workflow toward the correct next QA skill based on the user’s wording and any referenced paths.

Frequently Asked Questions about sumo-qa-deciding-approach

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

FAQPage Schema
How do I decide the right QA strategy before writing tests for a new change?

Determining the right QA strategy involves classifying your change intent, inferring its test-writing shape, and enforcing reachability gates before routing to scaffolding or planning. This prevents wasted effort by selecting the correct approach upfront.

What is a reachability gate in test planning and when is it needed?

A reachability gate in test planning checks for orphan or deletion conditions before choosing any test-writing approach. It is needed when classifying change shapes to enforce routing discipline and prevent invalid test generation.

When should I recommend removing tests instead of writing new ones during code review?

Recommending removal of tests occurs when the QA classification triggers a STOP condition, routing to a no-tests-recommended or recommend-removal approach based on the change intent and reachability gate results.

Can I use test-driven development scaffolding for repo-wide changes?

TDD scaffolding applies to specific change shapes determined by the QA router. The router classifies whether your request is single-change, repo-wide, or docs/config, ensuring scaffolding only triggers for appropriate test-writing scenarios.

What is the best way to route QA workflows for mutation testing and regression strengthening?

The best way to route QA workflows is by loading classification and approach catalogues, then returning a routing payload with the correct next action skill. This handles mutation testing and regression strengthening by matching your change intent to the right sub-skill.

How do STOP conditions work when no tests are recommended for a change?

STOP conditions trigger when classification yields no-tests-recommended or recommend-removal, returning a routing payload with rationale but no sub-skill handoff. This prevents unnecessary test scaffolding for changes that do not require testing.