skills.rest
sliec
sliecCOMMUNITY
Explore all skills by this author →
GitHub

prd-auto-test-loop

Generates versioned test plans and reports from PRD acceptance criteria.

Updated Jun 22, 2026
Search Tags:#automation#regression testing#e2e testing#prd#test plan#release gate#test report
One-click install
npx skills add https://github.com/sliec/shared-files --skill prd-auto-test-loop-sliec
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: prd-auto-test-loop
Source: https://github.com/sliec/shared-files/tree/main/skills/prd-auto-test-loop
Command: npx skills add https://github.com/sliec/shared-files --skill prd-auto-test-loop-sliec

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? It prevents teams from rebuilding entire test suites for every PRD release by enforcing a shared regression baseline plus per-version incremental tests, with standardized TEST_PLAN and TEST_REPORT outputs. ## Core Features & Use Cases - Test Layering from Acceptance Criteria: Maps PRD acceptance criteria to Unit, Integration, and E2E levels, and marks each as automatable (A), manual (H), or hybrid (A+H). - Versioned Test Artifacts: Creates a per-version directory with TEST_PLAN.md and TEST_REPORT.md templates, keeping each PRD release's coverage traceable. - AI Self-Test Loop: Runs tests in a fixed Unit → Integration → E2E order, fixes root causes on failure with up to 3 retry rounds, and records blockers instead of claiming false passes. - Use Case: When shipping PRD V0.6, generate an incremental test plan that reuses the main regression suite, add only new/changed cases, execute the self-test loop, and produce a PASS/FAIL release gate report. ## Quick Start Use the prd-auto-test-loop skill to create a versioned test plan and test report for PRD version V0.6 based on its acceptance criteria.

Frequently Asked Questions about prd-auto-test-loop

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

FAQPage Schema
How do I create a test plan from a PRD?▼

Extract the PRD's version goals, test scope, non-scope, and acceptance criteria, then map each criterion to Unit, Integration, or E2E levels. The skill generates a TEST_PLAN.md template with completion gates, case lists, and execution order.

How to decide which tests to automate versus test manually?▼

Mark each acceptance criterion as A (fully automatable), H (manual only), or A+H (hybrid). Functional correctness, regression paths, and deterministic edge cases suit automation; usability, visual quality, and business judgment need manual verification.

Should I rebuild the entire test suite for each new PRD version?▼

No. Maintain one shared main regression suite across versions and add only incremental cases for new or changed PRD functionality. Each version gets its own TEST_PLAN and TEST_REPORT under a versioned directory.

What happens when automated tests fail during the self-test loop?▼

Failures are fixed by root cause and re-run, with a maximum of three repair rounds. If blocked by environment issues, the blocker is explicitly recorded and the run is not claimed as passed.

What are the limitations of AI-driven automated testing?▼

Automation cannot fully cover usability, visual consistency, business strategy reasonableness, or open-ended exploratory paths. These items must be listed as residual risks in the TEST_REPORT and verified manually before release.

Related Skills

Explore More Product & Management Skills→
denish12denish12

pm_prd

Draft a complete PRD with structured sections and testable acceptance criteria.

Community
Advanced
mlinnenmlinnen
1

release-validation-triage

Triage release readiness by re-running failing tests in isolation.

Community
Intermediate
mahg-esmahg-es
3

uat-review

Validate UAT packages for requirements coverage, traceability, and readiness.

Official
Intermediate
pantheon-orgpantheon-org
9

bdd-testing

Create and run Cucumber BDD tests using Gherkin syntax and feature files.

Official
Advanced