next-qa-idea

Identifies the highest-value untested behavior in a codebase and files one locked qa issue specifying the test.

5.4k|571|Updated Nov 1, 2025
One-click install
npx skills add https://github.com/breaking-brake/cc-wf-studio --skill next-qa-idea
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: next-qa-idea
Source: https://github.com/breaking-brake/cc-wf-studio/tree/main/.claude/skills/next-qa-idea
Command: npx skills add https://github.com/breaking-brake/cc-wf-studio --skill next-qa-idea

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

QA backlogs go stale or get filled with low-value test ideas when ideation and implementation are not paced together. This Skill runs one unattended ideation iteration of a quality-assurance loop: it finds the highest-value untested behavior in the repository, judges it against a strict QA value bar, and files exactly one locked qa issue that an implementation skill can build from later.

Core Features & Use Cases

  • Gap discovery: Scans open bug issues, recent commits on main, pure logic in core packages, and weak spots in the existing test suite to find behaviors that would break silently.
  • Strict value bar: Rejects proposals that do not fit the assurance-map suites, are non-deterministic, cannot ship in one iteration, or require editing product source code.
  • Queue back-pressure: Files nothing when 3 or more qa issues are already open, keeping ideation paced with implementation.
  • Locked issue specs: Creates and immediately locks the issue so the spec stays owner-authored, including protection value, exact target file and function, cases to cover, and blocking dependencies.
  • Use Case: Run it on a schedule alongside a next-qa implementation skill so the QA backlog refills itself with verified, high-value test specs without any human triage.

Quick Start

Ask the agent to run one QA ideation iteration and file the single best test proposal as a locked qa issue.

Frequently Asked Questions about next-qa-idea

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

FAQPage Schema
How do I automate QA test idea generation for a codebase?

Run one ideation iteration that scans open bug issues, recent commits, and untested pure logic, then verifies the gap in the code before filing a single locked qa issue. The issue body contains the target function, cases to cover, and blocking dependencies so an implementation pass can build from it directly.

How to keep a QA backlog from overflowing with stale test ideas?

Apply queue back-pressure: file nothing when 3 or more qa issues are already open, and cap output at one issue per run. This keeps ideation paced with implementation so specs do not rot as the code changes.

What makes a test proposal worth filing as an issue?

It must protect a user-facing behavior, catch a plausible regression, be deterministic with no network or wall-clock dependence, ship in one implementation iteration, and be testable without editing product source code. Proposals failing any criterion are rejected.

Does this skill write or modify any tests or code?

No. It is strictly read-only toward the repository: it never commits, branches, opens PRs, or edits files. Its only write operation is creating and locking a qa issue; a separate implementation skill writes the actual tests.

Why lock the GitHub issue immediately after creating it?

Locking restricts comments to collaborators, so the test spec stays authored by the repository owner and the automation loop. Outside comments cannot steer the spec, while the owner can still comment to give feedback or close the issue to veto it.