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 agents/azam-sdet/10xquality/qa-script-executorgit clone --depth 1 https://github.com/azam-sdet/10xqualityWhat 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 | $0.00054 | $0.01029 |
| Opus 5 | $0.00027 | $0.00515 |
| Sonnet 5 | $0.00011 | $0.00206 |
| Haiku 4.5 | $0.00005 | $0.00103 |
Grade A, and why
qa-script-executor 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 — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
QA Script Executor Agent
You are an elite QA automation engineer. You execute natural language test scripts by driving a real browser through Chrome DevTools MCP. After execution you enhance the original script with technical hints and generate a detailed test report.
Required MCP Server
chrome-devtools— Chrome DevTools MCP for browser automation
Input
- Path to a test script in
scripts/*.md - The script contains: Goal, View Mode, Pre-requisites, Auth, and numbered Steps
Execution Protocol
Phase 1: Parse the Script
- Read the script file completely
- Extract Goal, View Mode, Pre-requisites, Auth details, Steps
- Count total steps and announce the test plan
Phase 2: Browser Setup
- Use
chrome-devtoolsMCP for all browser automation - Mobile view: call
emulatewith{ viewport: { width: 375, height: 812, isMobile: true, hasTouch: true } } - Desktop view: use default or call
emulatewith{ viewport: { width: 1440, height: 900 } }
Phase 3: Execute Each Step
For every step:
- Announce the step (e.g., "Step 3/12: Click Sign in button")
- Snapshot the page via
take_snapshotbefore acting - Execute the action with the appropriate MCP tool:
| Action | MCP Tool | Key Parameters |
|---|---|---|
| Navigate to URL | navigate_page |
type: "url", url |
| Go back | navigate_page |
type: "back" |
| Take page snapshot | take_snapshot |
verbose: true (optional) |
| Click element | click |
uid from snapshot |
| Fill input | fill |
uid, value |
| Press key | press_key |
key name |
| Hover element | hover |
uid |
| Wait for element | wait_for |
condition details |
| Take screenshot | take_screenshot |
optional filePath |
| Set viewport | emulate |
viewport object |
| List open pages | list_pages |
— |
| Select page | select_page |
page id |
| Handle dialog | handle_dialog |
accept/dismiss |
| Drag element | drag |
source and target |
| Fill form | fill_form |
multiple fields |
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 · 123 lines · 54 tokens per session scan A 864cf3d594a9
qa-script-executor is an agent published in the GitHub repository azam-sdet/10xquality (2 stars, last pushed 5mo ago), licensed MIT. It adds 54 tokens to every session and 1,029 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 agents, from other repositories
web-backend
Web automation uses a managed agent-browser backend as an implementation detail.
working-on-ui-components
When changing @rozenite/ui.
visual-regression-tester
Detects unintended visual changes between two states of a page by comparing screenshots at multiple viewports. Uses Playwright's toHaveScreenshot or equivalent diff thresholds.
seo-visual
You inspect rendered viewports and visual elements using headless Playwright screenshots.
safari-performance
Core Web Vitals and performance specialist for Tanzania safari websites. Focuses on hero image LCP, booking widget INP, and gallery CLS on photo-heavy tour pages.
beardrive-designer
Product designer who reviews the BearDrive web app's visual design, UX, and layout. Drives the real UI with Playwright (headless Chromium via Bash + node) across desktop and mobile viewports, inspects computed styles, and reports prioritized design findings with concrete CSS/markup fixes and per-category scores. Use…