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.
git clone --depth 1 https://github.com/JohnWayneeee/casely-qa-skillWrote 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/commands/johnwayneeee/casely-qa-skill/run)<a href="https://agentmods.dev/commands/johnwayneeee/casely-qa-skill/run"><img src="https://agentmods.dev/badge/commands/johnwayneeee/casely-qa-skill/run/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/commands/johnwayneeee/casely-qa-skill/run"><img src="https://agentmods.dev/badge/commands/johnwayneeee/casely-qa-skill/run.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00023 | $0.00679 |
| Opus 5.5 | $0.00009 | $0.00272 |
| Sonnet 5.5 | $0.00005 | $0.00136 |
| Haiku 4.5 | $0.00002 | $0.00068 |
Grade A, and why
run 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 yesterday.
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 — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Run the test cases in a browser
Arguments: $ARGUMENTS
Walk the test cases Casely exported against a running web app and report what happened. You are the tester at the keyboard: follow each case's steps literally, observe, and record. Do not fix the app and do not rewrite the cases.
Before the first case
- Playwright MCP must be connected. Look for
browser_*tools from the Playwright MCP server. If none are available, stop and tell the user to run, once:claude mcp add playwright -- npx @playwright/mcp@latest, then restart the session. Do not fall back to writing a Playwright script. - Base URL. Take it from the arguments. If missing, ask. Never guess a production URL; if the user gives one, confirm that running write actions (sign-ups, payments, deletes) there is intended.
- The cases. Take the path from the arguments, else
exports/all_test_cases.xlsx, else the Markdown files inresults/. Read an.xlsxwithopenpyxl, never by eye. - Pick what runs. UI cases only. List the API cases (those also in the Postman collection) as "run with Newman" and skip them. Show the user the count and the first few IDs, and ask for a scope if there are more than 15 cases.
- Test data. If a case needs credentials or data the case does not carry, ask once for all of it before starting. Never invent a password or a card number.
Each case
- Reset to a clean state: open the base URL in a fresh page; sign in again when the case's preconditions say so.
- Satisfy the preconditions. If one cannot be met (no such user, feature flag off), mark the case Blocked with the reason and move on.
- Perform each step with the Playwright tools. Take a snapshot after the step that carries the expected result.
- Compare what the page shows with the expected result, literally. Close enough is a fail with a note, not a pass.
- Record: ID, result (Pass, Fail, Blocked), and for a fail what the page showed
instead and at which step. Save a screenshot for every fail and block under
run/screenshots/<ID>.png.
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.
- yesterday First seen · 53 lines · 23 tokens per session scan A 44d2e8d53d4f
run is a command published in the GitHub repository JohnWayneeee/casely-qa-skill (11 stars, last pushed yesterday), licensed MIT. It adds 23 tokens to every session and 679 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-10-26.
Other commands, from other repositories
verify-pr
Verify a PR's frontend changes through browser automation.
record
Record browser actions into a test definition.
tester
Run the E2E tester workflow from .cursor/skills/tester/SKILL.md.
expect
Diff-aware AI browser testing — reads the git diff, maps changes to affected pages via the route map, generates a targeted test plan, and executes it via agent-browser (Rust daemon + CDP, ARIA-tree-first) with pass/fail reporting. Use when testing UI changes, verifying PRs before merge, or running regression checks on…
webapp-testing
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.
record
Record a browser walkthrough of a URL using Antigravity (agy). Generates .webm video, screenshots, and a report. Auto-converts to MP4 if ffmpeg is available.