agentsclimarketplace

Real device qa

Skill meharajM/context.md/ContextEngine/skills/real-device-qa

Open-source local-first iOS context engine with a portable real-device QA Agent Skill for evidence-backed physical-device testing.

Install
npx -y skills add meharajM/context.md --skill real-device-qa

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Run deterministic, evidence-backed QA on a physical iOS device using WebDriverAgent. Use when validating app behavior, device-specific flows, or real-user journeys that cannot be proven in a simulator.

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

2.4 KB, as published. Nobody here has run it

Real Device QA Skill

Use this skill when the user asks to run QA on a physical iOS device, validate behavior from app-behaviour.md, or verify real-user flows.

Purpose

Run deterministic, evidence-backed real-device QA against the iOS app using WebDriverAgent (WDA), then report pass/fail by behavior area.

Preconditions

  1. Physical iPhone connected, unlocked, and trusted.
  2. App build installed (npm run ios -- --device "<Device Name>" --no-packager when needed).
  3. WDA reachable at http://<device-ip>:8100/status and returns ready: true.
  4. app-behaviour.md reviewed before execution.
  5. project-architecture.md reviewed before execution so QA is mapped to the actual runtime architecture and supported flows.

Workflow

  1. Verify device/tooling health:
    • xcrun xctrace list devices
    • mobilecli devices --include-offline
    • curl http://<device-ip>:8100/status
  2. Run scripted QA:
    • node scripts/real-device-qa.js --wda http://<device-ip>:8100 --bundle com.meharaj.contextengine
  3. Review generated evidence:
    • JSON summary in artifacts/real-device-qa/<timestamp>/qa_results.json
    • XML snapshots in the same directory.
  4. Map outcomes to app-behaviour.md implemented flows.
  5. Cross-check observed behavior against project-architecture.md when failures may reflect routing, queue, persistence, model, or native-boundary regressions.
  6. File issues for confirmed regressions with repro and evidence file paths.

Expected Outputs

  • Behavior matrix with PASS / FAIL / INCONCLUSIVE.
  • Concrete evidence files (XML/JSON).
  • GitHub issues for defects when available.

Maintenance

  • If QA work leads to a code change that alters architecture or a primary user flow, update project-architecture.md in the same task.

Failure Handling

  • If session creation fails with a lock error, request the device be unlocked and rerun.
  • If WDA is not reachable, treat as environment blocker and report separately from app failures.
  • Distinguish tooling blockers from product regressions in the final QA report.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.