agentmods.dev

antigravity-testing-kit GEMINI.md

A set of rules for Gemini, an AI coding agent, to run browser-based test automation with Playwright, a browser testing tool. It also restricts code-changing Git commands and requires a visible desktop browser during debugging.

In plain words
What is it for?
Use it when writing or debugging automated browser tests, checking desktop layouts, and investigating failures through a visible browser before running tests headlessly in continuous integration.
Why use it?
It prevents local code from being overwritten by commands such as pull or reset and makes UI failures easier to inspect. It also enforces a consistent debugging order.

Instructions file for Gemini CLI

131 1mo ago A scan 3,029 tokens original MIT
View the source file anhtester/antigravity-testing-kit
Install

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.

agentmods
npx agentmods add instructions/anhtester/antigravity-testing-kit/gemini-md
Clone the repo
git clone --depth 1 https://github.com/anhtester/antigravity-testing-kit

Made for: Gemini CLI.

Per session 3,029 This file is loaded in full into every session.
When invoked 3,029 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.03029 $0.03029
Opus 5 $0.01515 $0.01515
Sonnet 5 $0.00606 $0.00606
Haiku 4.5 $0.00303 $0.00303

Measured 2d ago against content hash d3b47b221ab9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

antigravity-testing-kit GEMINI.md 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 2d 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.

GEMINI.md · 244 lines

How it starts

The opening of the file, as written. The whole thing — 244 lines — stays where its author put it; the contents beside it link to each section on GitHub.

GEMINI AI - GLOBAL AUTOMATION AGENT RULES

Scope: Áp dụng cho mọi tác vụ Test Automation do Gemini (Antigravity) hoạt động trong dự án này. Mục tiêu: Sinh ra test scripts hiệu quả, ổn định – dễ debug – dễ scale – CI friendly.

Git Pull Restriction Rule

  • Tuyệt đối KHÔNG dùng lệnh GIT làm thay đổi trạng thái code (như git pull, git checkout, git merge, git rebase, git reset) để lấy code hoặc thay đổi nhánh.
  • Vì có nhiều trường hợp code trên server chưa được cập nhật, việc pull code về sẽ ghi đè và làm sai toàn bộ phần code đang chỉnh sửa ở máy local.
  • Luôn giữ nguyên trạng thái code local hiện tại để làm việc.
  • Nếu cần file hoặc nội dung mới, hãy yêu cầu người dùng cung cấp thay vì tự ý dùng git.
  • Được phép dùng lệnh read-only: git status, git diff, git log — để kiểm tra trạng thái mà không thay đổi code.

Browser Rules (MANDATORY)

🖥️ Viewport & Mode

  • Tất cả UI debugging phải chạy với desktop viewport: 1920x1080
  • Bắt buộc mở browser thật khi debug (headed mode)
  • Headless mode chỉ được sử dụng sau khi test đã debug PASS trên UI
  • CI/CD pipeline được phép chạy headless mặc định

🔄 Thứ Tự Debug Bắt Buộc (Playwright MCP)

Khi dùng Playwright MCP để debug UI, LUÔN tuân theo thứ tự:

navigate → resize(1920×1080) → wait_for(page_load) → snapshot → interact → screenshot(on_fail)
  • KHÔNG gọi browser_navigate lại nếu đã đang ở đúng trang — tránh reload ngoài ý muốn
  • LUÔN gọi browser_resize(width=1920, height=1080) ngay sau browser_navigate
  • LUÔN verify page đã load xong trước khi lấy snapshot hoặc tương tác

📸 Screenshot & Snapshot

  • Dùng snapshot để phân tích DOM và xác định locator
  • Dùng screenshot để lưu bằng chứng khi test fail hoặc để báo cáo
  • Chụp screenshot ngay khi assertion fail để hỗ trợ truy vết lỗi
  • KHÔNG chụp screenshot tràn lan — chỉ khi cần thiết (fail / milestone quan trọng)

Tools

🛠️ Ưu Tiên Sử Dụng

Read the full file on GitHub · 244 lines

Changes

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.

  1. 2d ago First seen · 244 lines · 3,029 tokens per session scan A d3b47b221ab9

Subscribe to this mod's changes

antigravity-testing-kit GEMINI.md is an instructions file published in the GitHub repository anhtester/antigravity-testing-kit (131 stars, last pushed 1mo ago), licensed MIT. It adds 3,029 tokens to every session, about $0.0151 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.

Browser automationE2E testingTestinganhtesterantigravitykittest-automationtesting-tools
Related

Other instructions, from other repositories

keploy AGENTS.md

Instructions for keploy/keploy, covering agents.md, what this project is, build, run, lint, test, build the binary and standard build (what ci calls "build-no-race").

keploy/keploy · 4,945 tokens

agentic-playwright common-tasks.instructions.md

Instructions for idavidov13/agentic-playwright, covering ai prompt templates for agentic playwright, critical, instructions, phase 1: identify the task category and resolve paths and phase 2: select and customize the matching prompt template.

idavidov13/agentic-playwright · 5,109 tokens

agentic-playwright selectors.instructions.md

Instructions for idavidov13/agentic-playwright, covering selector strategy, critical, instructions, phase 1: open and authenticate and phase 2: explore like a user.

idavidov13/agentic-playwright · 2,174 tokens

agentic-playwright playwright-cli.instructions.md

Instructions for idavidov13/agentic-playwright, covering browser automation with playwright-cli, quick start, open new browser, navigate to a page and interact with the page using refs from the snapshot.

idavidov13/agentic-playwright · 1,693 tokens

setup-chromedriver CLAUDE.md

Instructions for nanasess/setup-chromedriver, covering claude.md, commands, build and development, running a single test and or.

nanasess/setup-chromedriver · 985 tokens

Graph-It-Live CLAUDE.md

Instructions for magic5644/Graph-It-Live, covering claude.md, commands, build, test and e2e (real vs code electron instance).

magic5644/Graph-It-Live · 1,643 tokens