simiancraft/simiancraft-skills / skills/ios-simulator-flow-evidence
ios-simulator-flow-evidence
VerifiedCapture proof of a driven iOS-simulator flow: vision-verifiable screenshots, video (mp4), and a manifest tying steps to artifacts, plus an honest list of what a simulator cannot do (deliver a push, use a real camera, place a call, fully observe the network). Produces artifacts to a contract that a separate GitHub-presentation skill consumes; this skill owns extraction, not publishing. Sits beside ios-simulator; project-agnostic. Use for "capture evidence of an iOS flow" or "record the simulator".
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 3.7 KB |
| references/artifact-contract.md | markdown | 2.4 KB |
| references/flow-manifest.md | markdown | 2.9 KB |
| references/screenshots.md | markdown | 2.7 KB |
| references/video.md | markdown | 4.1 KB |
references/screenshots.md
View on GitHub---
title: Screenshot evidence
summary: capturing one PNG per driven step and vision-verifying each as proof of a flow
status: complete
sources:
- "axe screenshot --help (PNG output, --output) and xcrun simctl io <udid> screenshot (PNG; see ios-simulator references/capture.md for --type and --mask)"
- https://developer.apple.com/documentation/xcode/capturing-screenshots-and-videos-from-simulator
---
# Screenshot evidence
Reference for **ios-simulator-flow-evidence**. A sequence of stills is the default, most
reviewable proof of a flow: one PNG per meaningful step, each verified to show the state it
claims. The capture mechanics are **ios-simulator** `references/capture.md`; this file is
the evidence discipline on top of them.
## One PNG per step
Drive the flow (see **ios-simulator** `references/driving.md`), and capture a still after
each step that changes the screen. Either tool writes a PNG:
```bash
axe screenshot --udid <udid> --output 01-launch.png
xcrun simctl io <udid> screenshot 02-signed-in.png
```
Both read the **device framebuffer**, so the Simulator window's size and position are
irrelevant; never resize or full-screen it to "fit" a shot (see `references/capture.md`).
## Freeze the status bar first
Before capturing, override the status bar so the clock, battery, and signal are fixed rather than
drifting between shots (the standard clean-screenshot setup): `xcrun simctl status_bar <udid>
override --time "9:41" --batteryState charged --batteryLevel 100` (see **ios-simulator**
`references/device-state.md`), and `status_bar <udid> clear` when done.
## Name shots so the sequence is the story
Use a zero-padded index plus a state label, so a directory listing reads as the flow:
```
01-launch.png
02-signed-in.png
03-detail-open.png
```
The index gives order; the label gives meaning. The manifest
(`references/flow-manifest.md`) ties each filename to the step and the assertion it proves.
## Vision-verify every shot
A screenshot is proof only once you have confirmed it shows the expected state. After each
capture, look at the PNG and check the concrete thing the step was supposed to produce (the
right screen, the expected row, the success toast); do not assume the tap landed. A green
run with a screenshot of the wrong screen is a false pass. Record the check in the manifest's
`verified` field.
## See also
- **ios-simulator** `references/capture.md`: the framebuffer, `--type`, and `--mask`.
- **ios-simulator** `references/driving.md`: driving the steps you are capturing.
- `references/flow-manifest.md`: tying each PNG to its step and assertion.
- `references/video.md`: when a continuous recording suits the flow better than stills.
- **ios-simulator** `references/device-state.md`: status-bar overrides for deterministic shots.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |