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 agentmods add skills/adambien/airails/web-system-testsnpx skills add AdamBien/airails --skill web-system-testsgit clone --depth 1 https://github.com/AdamBien/airailsWrote 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/adambien/airails/web-system-tests)<a href="https://agentmods.dev/skills/adambien/airails/web-system-tests"><img src="https://agentmods.dev/badge/skills/adambien/airails/web-system-tests.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00239 | $0.02762 |
| Opus 5 | $0.00120 | $0.01381 |
| Sonnet 5 | $0.00048 | $0.00552 |
| Haiku 4.5 | $0.00024 | $0.00276 |
Grade A, and why
web-system-tests 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 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.
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 — 212 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Write, run, and maintain browser-driven system tests for $ARGUMENTS. Apply all rules below strictly.
Composition
/system-testsowns the contract — black box, running system, coordinates as configuration, isolation, total verdict reporting. This skill is its web realization: layout, configuration, selector policy, test syntax, and the green oracle.- The stack skill owns the application under test —
/web-components(SPA),/web-sprinkles, or/web-static. /ears-testsowns spec-derived expectations: onetest.describeper requirement group, the literalRn.mid in the test title./bceowns where things live: one spec file per business component, named after it.
Versus the Chrome DevTools verification loop
/web-static drives the rendered page through Chrome DevTools MCP — console, accessibility
snapshot, viewport, Lighthouse, Baseline. That loop inspects the page as it is right now:
agent-driven, single-shot, leaving no artifact behind. This skill produces a committed suite
re-run by a machine on every change, in three engines, indefinitely.
Neither replaces the other. Rendering, accessibility tree, contrast, and Baseline compliance belong to the DevTools loop; user-visible behavior, and its survival across changes, belongs here. Where a project has both, green means both.
The public surface of a web UI is its accessibility tree
/system-tests permits assertions through the public surface only. For an HTTP service that surface
is the endpoint; for a web UI it is everything the user can perceive and operate — which is exactly
what the accessibility tree exposes. So the selector policy is a consequence of the contract, not a
style preference:
- Locate by role, label, text, or placeholder:
getByRole,getByLabel,getByText,getByPlaceholder. - CSS selectors, ids, class names,
nth-child, DOM shape, shadow-root piercing, and store internals are implementation. A test bound to them fails on a refactor that changed nothing a user can see — which is a false alarm, and false alarms are what kill suites.
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.
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.
- 5d ago First seen · 212 lines · 239 tokens per session scan A 6fc2b6b6f257
web-system-tests is a skill published in the GitHub repository AdamBien/airails (48 stars, last pushed yesterday), licensed MIT. It adds 239 tokens to every session and 2,762 once invoked, about $0.0012 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 2026-08-30.
Other skills, from other repositories
playwright-cli
Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.
smoke-test
Create a Mastra project using create-mastra and smoke test the studio in Chrome using Chrome MCP server.
loom-e2e-testing
End-to-end testing for web applications with Playwright, Cypress, Selenium, and Puppeteer.
pn-smoke-tests
Run smoke or E2E tests, debug failures, and verify fixes. Use for end-to-end verification before or after changes.
playwright-cli
Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.
qa-manual
Drives a local web app through Chrome MCP across the happy path plus 2-3 edge cases, capturing screenshots, console errors, and network failures to produce QA evidence. Use when manually QA'ing a web feature, testing a flow in the browser, exercising a change before claiming it works, or producing evidence for…