Testing Skill Discovery, putting testing into the delivery rhythm

Testing Skill Discovery, putting testing into the delivery rhythm

nao.deng ·

Map entry points, dependencies, environments, risks, existing tests, and questions in an unfamiliar project. The task sounds familiar, yet evidence and boundaries are easy to lose. Testing Skill Discovery starts there.

Awesome QA Skills organizes Skills by language and testing stage. The series overview covers repository structure and shared installation options; this guide stays with Testing Skill Discovery.

Read the source Skill first

The main prompt covers Routing decision rules (apply in order), Quality Bar, Gotchas, Pre-delivery checklist, Workflow. Those headings are navigation; the project artifacts still provide the facts.

The source directory contains 17 script entries. Start with Template conversion script.

Put the task on a timeline

Map entry points, dependencies, environments, risks, existing tests, and questions in an unfamiliar project

Workflow Skills become useful at handoffs. This is a planning sample; dates, owners, and evidence must come from the project.

StageActionEvidenceOwner
EntryConfirm scope and environmentRequirement version, build IDQA
ExecutionRun high-risk journeys firstRun record, defect linksQA and engineering
DecisionCheck exit criteriaResidual risk listRelease owner
HandoffRecord unfinished workOwner and due dateProject lead

If approval ownership or evidence is blank, the plan is not ready for a decision meeting.

Give the workflow a normal path and interruption paths

The working task is Map entry points, dependencies, environments, risks, existing tests, and questions in an unfamiliar project. Put it into the actual delivery cadence.

MomentTeam actionArtifactExit condition
Before entryLock requirement version, build, and ownerScope noteHigh-risk unknowns have owners
During executionRun delivery-blocking journeys firstRun record and defectsP0 results are reviewed
Before decisionCollect omissions and residual risksDecision packetApprover has read the risks
After handoffTrack deferred workOwner, date, linkFollow-up exists in the work system

When an environment arrives late, finish static review and data preparation. Repeated build failures are blockers with duration, not failed product tests. A late scope addition triggers a new look at Routing decision rules (apply in order) and Quality Bar; it does not silently stretch the old plan.

A clean handoff names the tested build, evidence-backed journeys, accepted risks, and the owner of the next action.

A prompt you can adapt

Replace the bracketed fields with project facts. Specific material leaves less room for guessing.

Use the discover-testing Skill.

Task: Map entry points, dependencies, environments, risks, existing tests, and questions in an unfamiliar project
Version and environment: [requirement / build / environment]
Inputs: [file paths or links]
Scope: [included and excluded journeys]
Constraints: [accounts, data, time, compliance]

List entry criteria, actions, evidence, owners, and exit criteria on a timeline. Track environment and build blockers separately from product failures.
Finish with open questions. Do not invent missing facts.

Use the first pass to inspect structure and gaps. Supply missing material before asking for the handoff-ready artifact.

Advanced use, from one call to a maintained flow

Assign a workflow_run_id and connect scope, build, owners, and handoff. The next Testing Skill Discovery run should load unfinished work instead of rediscovering it.

A three-Skill chain

discover-testingrequirements-analysistest-strategy

HandoffPayloadReceiver check
Upstream to discover-testingSource versions, scope, risks, open questionsTesting Skill Discovery staleness and conflicts
discover-testing to downstreamPrimary artifact, evidence index, unfinished workTesting Skill Discovery executability and owners
Feedback to discover-testingRuns, defects, new risksTesting Skill Discovery baseline and regression update

Do not paste three complete outputs into one large prompt. Give Testing Skill Discovery a structured summary and accessible source artifacts. It saves context and makes defects traceable.

Team gates

GateCheckFailure action
discover-testing inputVersion, environment, owner, accessible sourcesStop Testing Skill Discovery and list gaps
discover-testing artifactMaterial claims carry basis and statusReturn Testing Skill Discovery for evidence
discover-testing executionCommand, exit status, report are reproducibleClassify infrastructure or test failure
discover-testing decisionResidual risks have accepter and dateDo not enter the next stage

Review Testing Skill Discovery adoption, human edit rate, unsupported claims, and failure-to-diagnosis time each sprint. Record a baseline for several cycles before setting targets.

Keep the workflow out of ceremony land

Testing Skill Discovery should say who acts, when they act, which evidence they produce, and what allows the next stage to begin. Meeting names matter less than Routing decision rules (apply in order), ownership, due dates, and exit criteria. Every unfinished item needs an owner.

Install and invoke

Install the individual Skill. The series overview carries the longer installation explanation.

npx skills add https://github.com/naodeng/awesome-qa-skills/tree/main/skills/en/testing-workflows/discover-testing -g -a codex -y

Invoke it with “Use the discover-testing Skill,” then attach the real artifacts.

Two practical questions

When does Testing Skill Discovery begin?

Begin when entry criteria are met. At minimum, know the scope, build, environment, owner, and target date.

Can the Skill make a Go/No-Go decision?

No. It organizes evidence, risk, and unfinished work. The accountable project owner decides.

What if the plan becomes stale halfway through?

Record when it changed, affected scope, and the new owner. Keep the old plan for the retrospective.

Who maintains the workflow artifacts?

Stage owners update their evidence; the handoff owner checks links and unfinished items.

Run Testing Skill Discovery against one real artifact and keep the input, output, and review notes. The fragments here establish structure; project evidence must still come from the project.

References

Share