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 agentskillexchange/skills --skill author-and-run-regression-tests-with-agent-qagit clone --depth 1 https://github.com/agentskillexchange/skillsWrote 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/agentskillexchange/skills/author-and-run-regression-tests-with-agent-qa)<a href="https://agentmods.dev/skills/agentskillexchange/skills/author-and-run-regression-tests-with-agent-qa"><img src="https://agentmods.dev/badge/skills/agentskillexchange/skills/author-and-run-regression-tests-with-agent-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.
<a href="https://agentmods.dev/skills/agentskillexchange/skills/author-and-run-regression-tests-with-agent-qa"><img src="https://agentmods.dev/badge/skills/agentskillexchange/skills/author-and-run-regression-tests-with-agent-qa.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 5 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 28 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 29 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 35 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 43 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 50 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 | $0.00046 | $0.00679 |
| Opus 5 | $0.00023 | $0.00340 |
| Sonnet 5 | $0.00009 | $0.00136 |
| Haiku 4.5 | $0.00005 | $0.00068 |
Grade A, and why
Author and Run Regression Tests with Agent QA 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 8d 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 — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Author and Run Regression Tests with Agent QA
Use Agent QA when a coding task needs an application flow verified in a real browser or mobile environment. Agent QA stores tests as reviewable project files, accepts natural-language actions and assertions, and can retain scoped observations from earlier runs. Prefer its MCP tools when available: discover the workspace, inspect configuration, generate canonical IDs, validate test or suite definitions, and enqueue the run. Use the CLI as a fallback. When execution fails, inspect the recorded steps and evidence before changing the test; distinguish an application regression from an outdated instruction or an infrastructure failure. Agent QA can re-observe the interface and try another path during a run, but successful self-healing should still be reviewed before the learned path is trusted for later runs.
Agent QA is currently source-available under FSL-1.1-ALv2. Each release converts to Apache-2.0 after two years. Configured model, browser, or device providers may have separate costs.
Prerequisites
Install Agent QA in the project and initialize its workspace:
npm install -D agent-qa
npx agent-qa init
npx agent-qa install-browsers --chromium
For Android or iOS projects, install the required mobile drivers:
npx agent-qa install-mobile-drivers --all
Workflow
- Discover the Agent QA workspace and inspect its active targets, devices, and providers.
- Generate IDs with Agent QA tooling; never invent test, suite, hook, run, or observation IDs.
- Create or update the natural-language test, then validate its definition before saving or running it.
- Enqueue the test or suite through MCP. If MCP is unavailable, run a validated file with
npx agent-qa run <test-file>. - Review step evidence, healing attempts, and retained observations before deciding whether to fix the application, revise the test, or retry infrastructure.
Installation
Install or set up from the source-backed instructions:
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.
- 8d ago First seen · 59 lines · 46 tokens per session scan A 573645479ac7
Author and Run Regression Tests with Agent QA is a skill published in the GitHub repository agentskillexchange/skills (38 stars, last pushed today), licensed MIT. It adds 46 tokens to every session and 679 once invoked, about $0.0002 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 2026-09-03.
Other skills, from other repositories
ironbee-android-devtools-cli
CLI for driving an Android emulator (AVD) over adb. Use when the user needs to connect to an emulator (attach to a running one or boot/manage an AVD), tap/swipe/type on the UI, read the UI/accessibility tree, take screenshots or screen recordings, capture logcat, capture HTTP(S) requests in-process (Frida OkHttp hook…
ironbee-browser-devtools-cli
Automates browser interactions using Playwright for web testing, debugging, and automation. Use when the user needs to navigate websites, take screenshots, fill forms, click elements, extract page content (HTML/text), audit accessibility (ARIA/AX tree), measure Web Vitals performance, monitor console logs and HTTP…
browser-testing
Automated browser testing, interaction automation, and form testing. Use when the user needs to test web pages, automate browser interactions, fill forms, test validation, run multi-step wizards, or test login/signup flows.
mobile-testing
Automated Android app testing on an emulator — connect to an AVD, drive the UI (tap/swipe/type/keys/deep links), read the UI/accessibility tree, capture screenshots/recordings and logcat, capture/mock/intercept the app's HTTP(S) traffic in-process (Frida OkHttp hook — no proxy/CA, pinning bypassed) and pin a W3C trace…
visual-testing
Visual testing, UI verification, and design comparison using screenshots and Figma integration. Use when the user wants to verify UI appearance, compare with Figma designs, test responsive layouts, check for visual regressions, or validate design implementation.
backend-testing
Test backend services end-to-end. Use when the user wants to drive HTTP/gRPC/GraphQL/WebSocket endpoints against a real server, correlate requests with server logs (file/Docker/Kubernetes), verify database side-effects with snapshot/diff or watch-changes, replay captured curl/HAR requests, or seed and roll back…