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.
npx skills add tjdrhs90/rn-launch-harness --skill rn-harness-acceptancegit clone --depth 1 https://github.com/tjdrhs90/rn-launch-harnessWrote 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.
[](https://agentmods.dev/skills/tjdrhs90/rn-launch-harness/rn-harness-acceptance)<a href="https://agentmods.dev/skills/tjdrhs90/rn-launch-harness/rn-harness-acceptance"><img src="https://agentmods.dev/badge/skills/tjdrhs90/rn-launch-harness/rn-harness-acceptance/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.
<a href="https://agentmods.dev/skills/tjdrhs90/rn-launch-harness/rn-harness-acceptance"><img src="https://agentmods.dev/badge/skills/tjdrhs90/rn-launch-harness/rn-harness-acceptance.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 4 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 MCP Rug Pull · line 38 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 40 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 72 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 73 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00060 | $0.01400 |
| Opus 5 | $0.00030 | $0.00700 |
| Sonnet 5 | $0.00012 | $0.00280 |
| Haiku 4.5 | $0.00006 | $0.00140 |
Grade A, and why
rn-harness-acceptance scanned grade A with 0 findings 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 6d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 134 lines — stays where its author put it; the contents beside it link to each section on GitHub.
rn-harness-acceptance — Phase 7.5: User Acceptance Gate
The last checkpoint before irreversible work (store build → submission). The agent Evaluator (Phase 6) already judged that the app works; this phase asks a different question a machine can't answer: "is this the app the user actually wanted, and are they OK shipping it?"
Runs after AdMob (Phase 7) so the user reviews the real, ad-integrated app, and before Build (Phase 8) so nothing gets built or submitted without an explicit human yes.
Evaluator PASS → AdMob → [👤 THIS PHASE] → Build → Screenshots → Submit
Trigger
Called by the orchestrator as Phase 7.5. Skipped only if config.md → skip_acceptance: true (e.g. unattended CI) — and even then, log loudly that a human gate was bypassed.
Input
- Project code (AdMob integration complete)
docs/harness/plans/YYYY-MM-DD-prd.md(PRD — core user journeys)docs/harness/config.md
Two parts
Part 1: Automated E2E Smoke (objective floor)
Give the human a known-good starting point — don't ask them to approve an app that fails its own core flows.
Why Maestro: the screenshot phase already uses it, so it's a cheap add. Maestro drives the real app on a simulator/emulator via YAML flows. (Alternatives: Detox — RN-specific, heavier setup; Appium — cross-platform, heaviest. Default to Maestro here.)
-
Launch the app (AdMob = native module → Expo Go will crash; use a dev client / native run):
# iOS (macOS + Xcode) npx expo run:ios # Android (SDK + emulator) npx expo run:android -
Create/reuse E2E flows under
.maestro/— one flow per core user journey from the PRD. Keep them smoke-level (does the happy path work end to end), not exhaustive:# .maestro/smoke_core.yaml appId: com.company.app --- - launchApp - assertVisible: "<home screen anchor text>" - tapOn: "<primary CTA>" - assertVisible: "<expected next screen>" # ...one block per PRD journey (create, list, edit, navigate, etc.)
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.
- 6d ago Changed · +4 lines c5ce7766e7fb
- 10d ago First seen · 130 lines · 60 tokens per session scan A 439d4b165619
rn-harness-acceptance is a skill published in the GitHub repository tjdrhs90/rn-launch-harness (9 stars, last pushed 9d ago), licensed MIT. It adds 60 tokens to every session and 1,400 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
flutter-testing
A testing guide for Flutter mobile apps and NestJS backends used across multiple regions. It covers unit tests, screen tests, end-to-end tests, regional build variants called flavors, and rules for testing payments, tenants, and external services with test doubles.
ui-crawler-max
Autonomous UI crawler for iOS simulator apps — rapid, element-precise data collection. Use whenever the user says "crawl my app", "auto-test the UI", "click/tap through every screen", "monkey test", "explore the app and find errors/crashes", "collect screenshots of all screens", "smoke test the whole app", or asks for…
android-ui-journey-testing
XML-specified Android UI journey testing, interactive step execution, assertion verification, and JSON outcome reporting.
android_ui_verification
Automated end-to-end UI testing and verification on an Android Emulator using ADB.
solopi-ai
A command-line framework for testing Android apps and devices with SoloPi, including on-device or cloud AI decision models. It manages devices, test cases, recorded interactions, replays, performance history, and evidence.
dogfood
Systematically explore and test a mobile app on iOS/Android with agent-device to find bugs, UX issues, and other problems. Use when asked to dogfood, QA, exploratory test, find issues, bug hunt, or test this app on mobile.