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/liortesta/clawdagent/visual-verifygit clone --depth 1 https://github.com/liortesta/ClawdAgentWhat 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 | $0.00000 | $0.00446 |
| Opus 5 | $0.00000 | $0.00223 |
| Sonnet 5 | $0.00000 | $0.00089 |
| Haiku 4.5 | $0.00000 | $0.00045 |
Grade A, and why
visual-verify 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.
What it actually says
Visual Verification — Browser-Based UI Testing
Use Playwright MCP to visually verify the UI that was just built or modified.
Process:
1. Start Dev Server (background)
- Run dev server in background if not already running
- Wait for server to be ready (health check)
2. Navigate & Screenshot
- Open the target URL in browser
- Take a full-page screenshot as "current state"
- If a reference screenshot exists, compare them
3. Interactive Testing
- Click through the critical user flow
- Fill forms with test data
- Verify navigation works
- Check responsive behavior (desktop + mobile viewport)
4. Visual Checks
For each page/component:
- Layout matches expectations
- No overlapping elements
- Text is readable (no truncation)
- Colors and spacing are consistent
- Loading states work
- Error states display correctly
- Empty states are handled
5. Screenshot → Fix Loop
If issues found:
- Take screenshot of the issue
- Identify the CSS/HTML problem
- Fix the code
- Re-screenshot to verify fix
- Repeat until clean
Output Format:
## Visual Verification Report
### Pages Tested
| Page | URL | Status | Screenshot |
|------|-----|--------|-----------|
| Home | / | PASS/FAIL | [description] |
| Login | /login | PASS/FAIL | [description] |
### Issues Found
- [issue description — file:line — fix applied]
### Responsive Check
| Viewport | Status |
|----------|--------|
| Desktop (1920x1080) | PASS/FAIL |
| Tablet (768x1024) | PASS/FAIL |
| Mobile (375x667) | PASS/FAIL |
### Verdict: VISUAL OK / NEEDS FIXES
Rules:
- ALWAYS test at minimum 2 viewports (desktop + mobile)
- ALWAYS check error states, not just happy path
- If Playwright MCP is not available, report and suggest manual testing
$ARGUMENTS
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.
- 2d ago First seen · 69 lines · 0 tokens per session scan A ed0ebd7652fd
visual-verify is a command published in the GitHub repository liortesta/ClawdAgent (11 stars, last pushed 6d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 446 tokens. 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
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…
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.
stark-quality-gate
Use this command when a web UI should be maintained, released, compared across runs, or used as public proof. It turns visual QA into a repeatable local/CI gate instead of a one-off screenshot check.
open
SoDam-Design-Kit open 대시보드 — 검증 이력·판정서·스크린샷을 브라우저로 열람 + 재검증.
playwright-cli
Use this skill whenever the user wants to automate a browser, scrape web content, take screenshots or PDFs of pages, fill out forms, click through UI flows, or run end-to-end tests — without using an MCP server. This skill drives Playwright directly from the bashtool via Node.js scripts. Trigger whenever the user says…
qa
Dispatch a verifiable browser task to the qa-tester subagent. Use for regression checks, smoke tests, and any task with a clean pass/fail outcome. Supports an EXHAUSTIVE mode that verifies every control behind an honesty gate.