Guide · Agent skill
How to use the breakdown-test agent skill for test planning
Updated · written with AI from the project's own documentation and checked against our catalogue data
The breakdown-test skill is a set of instructions that makes your AI agent behave like a senior QA engineer. Given a feature's requirements and technical plans, it drafts a test strategy, a checklist of test issues and estimated test tasks for GitHub project management. This guide explains what the skill does, how to install it, how to trigger it and where its output needs a human review.
Quick start
- 1Download the ZIP. Get the breakdown-test ZIP from the directory page and unzip it. It contains SKILL.md.
- 2Place the skill folder. Move the folder into ~/.claude/skills/ for personal use or into a project's .claude/skills/ folder for one project. Or upload it under Settings → Capabilities → Skills in the Claude apps.
- 3Document the feature. Make sure the feature PRD, technical breakdown, implementation plan and GitHub project plan exist in the expected docs/ways-of-work/plan folder layout.
- 4Ask for a test plan. Send a natural request naming the epic and feature, such as asking for a test strategy and test issues checklist.
- 5Check the three output files. Open test-strategy.md, test-issues-checklist.md and qa-plan.md in the feature folder.
- 6Review with a human tester. Compare scope, risks and estimates with the real system and adjust them before creating GitHub issues.
What the breakdown-test skill makes the agent do
When installed, the agent works from a feature's requirements and technical plans and produces structured test planning documents. It applies ISTQB test design techniques such as equivalence partitioning and boundary value analysis, and it uses the ISO 25010 quality characteristics to organise quality validation.
The skill covers three outputs: a test strategy, a test issues checklist, and a quality assurance plan. The checklist covers unit, integration, end-to-end, performance, security and accessibility test issues, plus estimated test tasks.
- Test strategy: scope, quality objectives, risks, test design techniques and environment and data plans
- Test issues checklist: test-level issues, test type priorities, dependencies, coverage targets
- Task breakdown: test implementation tasks with story-point estimates for GitHub Projects
- Quality assurance plan: entry and exit criteria for quality gates
Install the breakdown-test skill
Download the ZIP for breakdown-test from the directory page and unzip it. The skill is a folder that contains a single file, SKILL.md.
For a personal install, place the unzipped folder in ~/.claude/skills/. To share it with one project, place it in that project's .claude/skills/ folder instead.
If you use the Claude apps, you can upload the skill under Settings → Capabilities → Skills rather than copying files.
~/.claude/skills/Prepare the feature documents first
The skill expects your feature to already be documented. It reads four files in a fixed folder layout, so have them in place before you ask for a test plan.
The skill writes its outputs into the same feature folder, so the plan sits next to the documents it was built from.
- Feature PRD: /docs/ways-of-work/plan/{epic-name}/{feature-name}.md
- Technical breakdown: /docs/ways-of-work/plan/{epic-name}/{feature-name}/technical-breakdown.md
- Implementation plan: /docs/ways-of-work/plan/{epic-name}/{feature-name}/implementation-plan.md
- GitHub project plan: /docs/ways-of-work/plan/{epic-name}/{feature-name}/project-plan.md
Trigger the skill with a natural request
You do not need a special command. Ask for test planning for a named feature and point the agent at its documents. Name the epic and feature so the agent can find the right folder.
The skill then writes test-strategy.md, test-issues-checklist.md and qa-plan.md into the feature folder.
- Create a test strategy and test issues checklist for the feature in epic checkout, feature guest-payments, using the existing PRD and plan files.
- Draft the QA plan and test task estimates for the guest-payments feature.
- Break down the testing work for my feature so it can go into GitHub Projects as issues.
Review the output before you use it
The skill only writes planning text. The file itself contains no scripts and does not connect to the internet, so nothing runs on your machine when you use it. The generated plan is a first draft, and the skill's own description says a human tester must review it.
Check the test design techniques, risks and estimates against how your real system behaves. Treat the story-point ranges as starting points for team discussion, not commitments.
- Confirm the scope matches the real feature and its current code.
- Replace generic risks with ones your team has actually seen.
- Adjust estimates using your team's own velocity and experience.
What people use it for
- Drafting a first-pass test strategy for a feature that already has a PRD and implementation plan.
- Turning test work into trackable GitHub issues for unit, integration, end-to-end, performance, security and accessibility testing.
- Estimating test tasks in story points for sprint planning in GitHub Projects.
- Learning how ISTQB test design techniques and ISO 25010 characteristics apply to a real feature.
Frequently asked questions
How do I install the breakdown-test skill?
Download the ZIP from the directory page and unzip it. Place the folder in ~/.claude/skills/ for personal use or in a project's .claude/skills/ folder. In the Claude apps, you can instead upload it under Settings → Capabilities → Skills.
What does the breakdown-test skill require?
The source lists no requirements. It does run scripts or use the network, according to its listing. It does expect your feature to already have a PRD, technical breakdown, implementation plan and GitHub project plan in the expected folder layout.
Is the breakdown-test skill free?
The skill is published under the MIT licence in the github/awesome-copilot repository. This page does not list any price.
Does breakdown-test run scripts or connect to the internet?
No. The skill file contains no scripts and does not connect to the internet, and it only writes planning text. Your agent still generates the documents, so review what it writes.
Can I use breakdown-test without existing feature documents?
It is not a good fit. The prompt expects a PRD and implementation files to exist, and without them the output will be generic and may not match your real system.
Can this replace a human QA sign-off?
No. The generated plan is an AI draft that a human tester must review. It is not suitable for certified or compliance-grade QA sign-off.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
Use breakdown-test if you are a developer or QA lead with a documented feature who wants a structured first draft of a test plan and GitHub test tasks. Pick something else if your project has no feature documents, or if you need certified or compliance-grade QA sign-off, which this AI draft cannot provide.
Download the skill