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/sagaaidev/mcp-ahrefs/test-ui-subprocessgit clone --depth 1 https://github.com/SAGAAIDEV/mcp-ahrefsWrote 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/sagaaidev/mcp-ahrefs/test-ui-subprocess)<a href="https://agentmods.dev/commands/sagaaidev/mcp-ahrefs/test-ui-subprocess"><img src="https://agentmods.dev/badge/commands/sagaaidev/mcp-ahrefs/test-ui-subprocess.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 5.1 | $0.00010 | $0.00822 |
| Opus 5 | $0.00005 | $0.00411 |
| Sonnet 5 | $0.00002 | $0.00164 |
| Haiku 4.5 | $0.00001 | $0.00082 |
Grade A, and why
test-ui-subprocess 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.
How it starts
The opening of the file, as written. The whole thing — 134 lines — stays where its author put it; the contents beside it link to each section on GitHub.
I'll execute comprehensive UI testing using a Claude subprocess for autonomous test execution.
Step 1: Prepare Test Context
Gather Information
!echo "Current branch: $(git branch --show-current)" !echo "Application URL: [Will need to determine from config or ask user]"
Verify Puppeteer MCP Availability
I'll launch a subprocess that can use the Puppeteer MCP tools for UI testing.
Step 2: Determine Test Scope
Based on the feature and JIRA issue, I'll identify:
- User flows to test
- Critical UI interactions
- Visual validation requirements
- Edge cases to explore
Step 3: Launch Subprocess Testing
I'll execute the UI tests using a Claude subprocess:
claude -p "You are testing a web application UI. Use the Puppeteer MCP tools to:
1. Navigate to [APPLICATION_URL]
2. Test the following user flow: [SPECIFIC_FLOW_DESCRIPTION]
3. Take screenshots at key points:
- Initial page load
- After each major interaction
- Final state
- Any error states encountered
4. Verify the following:
- All interactive elements work correctly
- Form validations trigger appropriately
- Navigation flows as expected
- Visual elements render correctly
- Responsive design works on different viewports
5. Test edge cases:
- Invalid inputs
- Rapid clicking
- Browser back/forward navigation
- Session timeout handling
6. Document your findings including:
- Steps performed
- Any errors or unexpected behaviors
- Performance observations
- Accessibility issues noticed
Use viewport sizes:
- Desktop: 1920x1080
- Tablet: 768x1024
- Mobile: 375x667
Start with headless: false to observe the test execution." --dangerously-skip-permissions
Step 4: Process Results
After subprocess execution:
- Review the test output
- Analyze screenshots captured
- Identify any failures or issues
- Create summary of test results
Step 5: Create Test Report
Generate a comprehensive test report including:
- Test scenarios executed
- Pass/fail status
- Screenshots of key states
- Performance metrics
- Accessibility findings
- Recommendations for fixes
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 · 134 lines · 10 tokens per session scan A 2b6f9f2a6f75
test-ui-subprocess is a command published in the GitHub repository SAGAAIDEV/mcp-ahrefs (0 stars, last pushed 1y ago), licensed MIT. It adds 10 tokens to every session and 822 once invoked, about $0.0001 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 commands, from other repositories
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…
auto-verify
프론트엔드 UX 검증 — Playwright 기반 비주얼 검증을 실행합니다.
buildable-preview
Render the running prototype in a headless browser, screenshot it, and catch runtime/visual errors.
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.
vibereview
Legacy alias for the focused Randroid VibeReview skill.