ios-device-qa

ios-device-qa is a skill for Claude Code, Codex from OutlineDriven/odin-claude-plugin. It costs 55 tokens per session (2,016 once invoked), scanned A, original, Apache-2.0.

A workflow for testing an iOS app on a USB-connected physical iPhone through a debug bridge.

In plain words
What is it for?
It is for deploying a debug build to an iPhone, exercising the app, and capturing screenshots and session logs.
Why use it?
It verifies real-device interactions and produces a QA report instead of relying only on simulator testing.

Skill for Claude CodeCodex

Written for Claude Code and Codex: shipped in a Claude Code plugin, but also agents/openai.yaml present. Also seen: built for gstack.

Part of the odin-apple plugin — 5 skills shipped together

Good fit It is for deploying a debug build to an iPhone, exercising the app, and capturing screenshots and session logs.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/outlinedriven/odin-claude-plugin/ios-device-qa
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add OutlineDriven/odin-claude-plugin --skill ios-device-qa
Clone the repo
git clone --depth 1 https://github.com/OutlineDriven/odin-claude-plugin

Made for: Claude Code, Codex.

Or install odin-apple, the plugin that ships this one along with the rest of its 5 skills.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for ios-device-qa

README.md
[![agentmods](https://agentmods.dev/badge/skills/outlinedriven/odin-claude-plugin/ios-device-qa/github.svg)](https://agentmods.dev/skills/outlinedriven/odin-claude-plugin/ios-device-qa)
Your own site
<a href="https://agentmods.dev/skills/outlinedriven/odin-claude-plugin/ios-device-qa"><img src="https://agentmods.dev/badge/skills/outlinedriven/odin-claude-plugin/ios-device-qa/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for ios-device-qa

Your own site · 80×15
<a href="https://agentmods.dev/skills/outlinedriven/odin-claude-plugin/ios-device-qa"><img src="https://agentmods.dev/badge/skills/outlinedriven/odin-claude-plugin/ios-device-qa.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 55 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,016 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 2 findings, up to medium

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • medium Rogue Agent · line 38
    Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.
    Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
  • medium Rogue Agent · line 38
    Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.
    Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00055 $0.02016
Opus 5 $0.00028 $0.01008
Sonnet 5 $0.00011 $0.00403
Haiku 4.5 $0.00006 $0.00202

Measured 5d ago against content hash 589dd5755502, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

ios-device-qa scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- `curl: connection refused` to the daemon: the daemon crashed. Re-run `/ios-device-qa`; the spawn-race lock fails closed rather than double-binding.
plugins/odin-apple/skills/ios-device-qa/SKILL.md · 53 lines

How it starts

The opening of the file, as written. The whole thing — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.

iOS device QA

Contract

Field Bound contract
Trigger the user runs /ios-device-qa on an iOS app
Authority reversible-local: wire a Debug-only SPM bridge into the app build, deploy to a USB-connected iPhone, capture screenshots and session logs; remove the wiring before a Release build
Side effect debug-bridge wiring in the app build plus screenshots and session logs under ~/.gstack/
Done a device QA report with verified interactions is produced

Inputs

  • --source <dir>: the app source directory on disk. Required.
  • USB-connected, paired, trusted iPhone; macOS host with Xcode and Swift >= 5.9.
  • App manifest is SwiftPM (Package.swift); at least one file-scope @Observable class with // @Snapshotable-marked writable instance var fields of JSON-native types.
  • Optional --tailnet: expose the device to remote agents over Tailscale (requires tailscaled running and /var/run/tailscale.sock readable).
  • Optional --recording: render an "AGENT DEMO" watermark on screencasts.
  • Optional demo request ("demo", "show me", "I want to see it working"): drive every action through visible UI only.

Procedure

  1. Verify bridge compatibility before mutating the app. The generator supports file-scope @Observable classes only; ObservableObject, @StateObject, and other observation models produce no accessors. The documented dependency wiring assumes a SwiftPM manifest; for an .xcodeproj or .xcworkspace do not invent package or target wiring. If either requirement is unmet, stop the bridge bootstrap without modifying the app. Preserve any installed production or TestFlight build; when a separate QA build is needed, use an isolated bundle identifier and non-production entitlements so it coexists with the production app. Report fixture-driven state, provider UI, and actual external-provider success as distinct evidence tiers.
  2. Walk --source and identify every file-scope @Observable class. A property immediately preceded by the marker comment // @Snapshotable is snapshot-eligible: it must be a writable instance var with an explicit type and an internal or public setter, of a JSON-native scalar (String, Bool, integer widths, Float, Double, CGFloat), array, String-keyed dictionary, or Optional composition. Keys must be unique across observable classes. Stop with a source diagnostic instead of emitting a broken or lossy harness when any constraint is violated.
  3. Show the accessor list and ask the user whether to install the DebugBridge SPM dependency into Package.swift.
  4. Generate the local bridge package, typed accessors, and installed version marker with one deterministic command: gstack-ios-qa-regen --app-source "<source-dir>" --bridge-dir "<source-dir>/DebugBridge". The regenerator also removes the obsolete flat-file set from older ios-sync versions so no stale second harness remains in the app target.
  5. Add the generated DebugBridge local SPM dependency to Package.swift. It ships three Debug-config-only products: DebugBridgeCore (StateServer + bridge protocols), DebugBridgeTouch (in-process touch synthesis with iOS 18+ _UIHitTestContext SwiftUI hit-testing), and DebugBridgeUI (screenshot, elements, and mutation bridges). The app target depends on DebugBridgeUI with .when(configuration: .debug), transitively pulling in Core and Touch; Release builds refuse to link these targets.
  6. Wire the bridges from the @main App init, gated on #if DEBUG: call DebugBridgeUIWiring.installAll() before the StateServer opens its listener, then DebugBridgeManager.shared.start(appState: appState, register: AppStateAccessor.register) with the type discovered in step 2.
  7. Build and deploy: xcodebuild -scheme <SchemeName> -destination 'platform=iOS,id=<UDID>' build install. Launch via devicectl device process launch --device <UDID> --console <bundle-id> and capture the boot token printed to os_log on first run.
  8. Spawn the Mac-side daemon gstack-ios-qa-daemon. It acquires an exclusive flock on ~/.gstack/ios-qa-daemon.pid; a second invocation discovers the live daemon's port and connects rather than double-binding. The daemon immediately calls POST /auth/rotate on the iOS StateServer with a fresh in-memory-only token, so the boot token becomes useless ~5s later and anything scraping os_log past this point sees a dead credential. If a fresh daemon finds the app running after another daemon consumed that one-use token, it verifies the bundle owner, relaunches the target once, waits for the new token, verifies ownership again, then rotates. The app's StateServer binds loopback only (::1 + 127.0.0.1); tailnet ingress is exclusively the daemon's job.
  9. Run the vision-driven agent loop. Each iteration: GET /screenshot and save the PNG; GET /elements for the accessibility tree; GET /state/snapshot for only the // @Snapshotable fields; decide the next action against the test goal; POST /session/acquire to grab the device lock; execute POST /tap, /swipe, /type, or POST /state/<key> write; re-screenshot and compare; record a finding if buggy; POST /session/release once the iteration is done.
  10. In demo mode, drive every action through visible UI (/tap, /swipe, /type) and never use POST /state/* writes to skip steps; bump the screencap rate to 4fps so the recording shows each action as it happens. This override takes precedence over all other rules; viewers see the agent type every key and tap every button.
  11. In tailnet mode (--tailnet), the daemon also binds the Tailscale interface (never 0.0.0.0), fails closed if /var/run/tailscale.sock is missing, permission-denied, or returns an unparseable WhoIs response, and mints short-lived session tokens (default 1h, max 24h) for allowlisted remote identities. Capability tiers are ordered observe < interact < mutate < restore; granting restore implies all lower tiers. Observe covers /screenshot, /elements, GET /state/*, /healthz, /session/heartbeat; interact adds /tap, /swipe, /type; mutate adds POST /state/<key>; restore adds POST /state/restore. Every authenticated mutating tailnet request writes an audit row to ~/.gstack/security/ios-qa-audit.jsonl; rejections write to `attemp…
  12. Before a Release build, remove the DebugBridge SPM dependency and all #if DEBUG wiring. The structural guard (.when(configuration: .debug) plus a CI swift build -c release check) is the safety-critical path; this cleanup is a convenience flow.

Read the full file on GitHub · 53 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 5d ago First seen · 53 lines · 55 tokens per session scan A 589dd5755502

Subscribe to this mod's changes

ios-device-qa is a skill published in the GitHub repository OutlineDriven/odin-claude-plugin (35 stars, last pushed yesterday), licensed Apache-2.0. It adds 55 tokens to every session and 2,016 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-04.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens

next-partial-prefetching-adoption

Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…

vercel/next.js · 103 tokens

chronicle

Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…

microsoft/vscode · 72 tokens