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 commands/anhtester/claude-testing-kit/generate_automation_from_ui_flowgit clone --depth 1 https://github.com/anhtester/claude-testing-kitWrote 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/anhtester/claude-testing-kit/generate_automation_from_ui_flow)<a href="https://agentmods.dev/commands/anhtester/claude-testing-kit/generate_automation_from_ui_flow"><img src="https://agentmods.dev/badge/commands/anhtester/claude-testing-kit/generate_automation_from_ui_flow.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 | $0.00036 | $0.03117 |
| Opus 5 | $0.00018 | $0.01558 |
| Sonnet 5 | $0.00007 | $0.00623 |
| Haiku 4.5 | $0.00004 | $0.00312 |
Grade A, and why
generate_automation_from_ui_flow 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- generate_automation_from_ui_flow — 89% identical, 20 lines differ
How it starts
The opening of the file, as written. The whole thing — 288 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Workflow: Sinh Automation từ UI Flow
BẮT BUỘC (MANDATORY SKILL): Bạn PHẢI nạp và đọc kỹ nội dung của skill
skills-ui-debug-agent(tại.claude/skills/skills-ui-debug-agent/SKILL.md) trước khi bắt đầu. Ngoài ra tham khảo thêm skillskills-smart-locator-agentđể sinh locator ổn định vàskills-qa-automation-engineercho quy tắc automation chung.
Workflow này giúp agent thực thi trực tiếp một chuỗi thao tác UI trên browser thật, thu thập locators từ DOM thực tế, và sinh automation scripts hoàn chỉnh — tất cả trong một luồng tự động, không cần manual test case có sẵn.
⚠️ Nguyên tắc thực thi
- Tất cả output bằng Tiếng Việt
- TUYỆT ĐỐI KHÔNG ĐOÁN locator — phải lấy từ DOM thực tế bằng MCP browser tools
- Phải chạy từng bước UI trên browser thật trước khi sinh code
- Desktop viewport 1920×1080 cho tất cả UI debugging
- ⚠️ Rule E3: Khi test FAIL → tự đọc log → phân tích → sửa → chạy lại. KHÔNG hỏi user
Workflow này khác gì generate_automation_from_testcases?
from_testcases |
from_ui_flow (workflow này) |
|
|---|---|---|
| Input | File manual test cases có sẵn | Mô tả UI steps bằng lời hoặc URL + hành động |
| Approach | Đọc TC → inspect UI → sinh code | Chạy thật trên browser → thu thập locator → sinh code |
| Khi nào dùng | Đã có test case document | Chưa có TC, chỉ biết "vào trang này, click cái kia" |
Input cần thu thập
Agent cần ít nhất 1 trong các input sau từ user:
| Input | Ví dụ | Độ ưu tiên |
|---|---|---|
| URL + UI steps mô tả | "Vào https://example.com, login, tạo user mới" | ⭐ Phổ biến nhất |
| URL + recording/video | User cung cấp video thao tác | Tùy chọn |
| URL + screenshots | User cung cấp ảnh chụp từng bước | Tùy chọn |
| Chỉ URL | "Automate login flow của trang này" | Agent tự khám phá |
Nếu user chưa cung cấp đủ → hỏi:
- URL ứng dụng?
- Mô tả flow cần automate (từng bước)?
- Credentials nếu cần đăng nhập?
- Framework mong muốn? (mặc định: Playwright + TypeScript)
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 · 288 lines · 36 tokens per session scan A 2982755f589e
generate_automation_from_ui_flow is a command published in the GitHub repository anhtester/claude-testing-kit (50 stars, last pushed 27d ago), licensed MIT. It adds 36 tokens to every session and 3,117 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 2026-08-30.
Other commands, from other repositories
generate_automation_from_testcases
Convert manual test cases into automation scripts autonomously using the 6-step AI-RBT Framework via Antigravity Capabilities.
generate_automation_from_ui_flow
Thực thi UI flow trực tiếp trên browser, thu thập locators từ DOM thực tế, và sinh automation scripts. Hỗ trợ Playwright, Selenium, Appium.
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…
buildable-preview
Render the running prototype in a headless browser, screenshot it, and catch runtime/visual errors.
auto-verify
프론트엔드 UX 검증 — Playwright 기반 비주얼 검증을 실행합니다.
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.