stage-qa

Runs incremental batch-scoped quality checks after code-touching apply batches in SDD v2.

Updated Apr 2, 2026
One-click install
npx skills add https://github.com/nico0695/ai-tools --skill stage-qa-nico0695
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: stage-qa
Source: https://github.com/nico0695/ai-tools/tree/main/sdd/sdd-v2/skills/stage-qa
Command: npx skills add https://github.com/nico0695/ai-tools --skill stage-qa-nico0695

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? It catches regressions early during spec-driven development by running fast, targeted validation after relevant code-touching batches, without pretending that final verification has already happened. ## Core Features & Use Cases - Incremental Validation: Runs focused smoke, regression, lint, or type checks scoped to the latest code-touching batch instead of a full test matrix. - Checkpoint Gating: Requires explicit user confirmation before running, with options to run now, defer, or stop and revise. - Evidence Recording: Writes QA outcomes (passed, warning, failed) and check evidence into apply-progress.md and state.yaml. - Use Case: After a batch marked "QA Checkpoint: end-of-feature" completes, run incremental QA to confirm the change is safe to continue before the next batch executes. ## Quick Start Run incremental QA on the latest code-touching batch and record the results in apply-progress.md.

Frequently Asked Questions about stage-qa

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

FAQPage Schema
How do I run incremental QA between implementation batches?▼

Trigger incremental QA after a code-touching batch marked end-of-feature or critical-point, or on explicit request. The phase selects focused smoke and regression checks from the batch validation note, runs them, and records the outcome in apply-progress.md.

What is the difference between incremental QA and final verification?▼

Incremental QA is batch-scoped, selective, and designed for early alerts during execution. Final verification is change-wide, evidence-heavy, and mandatory when code changed; passing incremental QA never satisfies final verification requirements.

When should incremental QA not be used?▼

Skip it for purely documentary or planning batches, as a mandatory gate after every small code change, or as a substitute for final verification. It only applies to relevant code-touching batches or explicit user requests.

Does incremental QA run automatically after each batch?▼

No, it is confirm-before-run. The orchestrator opens a checkpoint summarizing the batch, changed scope, and planned checks, and the user chooses to run now, defer, or stop and revise.

What happens when incremental QA returns a warning?▼

A warning means non-blocking issues or uncertainty remain. The result is recorded in apply-progress.md, open risks may be added to state.yaml, and the run is reported with partial status rather than success.