ui-visual-validator

ui-visual-validator is an agent for Claude Code from NOMARJ/sigil. It costs 36 tokens per session (697 once invoked), scanned A, original, Apache-2.0.

A visual UI checker that uses screenshots to determine whether an interface change achieved its intended result.

In plain words
What is it for?
Use it to validate visual fixes, compare screenshots with design goals, and check that UI changes are complete.
Why use it?
It catches layout, spacing, sizing, alignment, and other visible problems that code inspection alone may miss.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md). Also seen: model in frontmatter.

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 agents/nomarj/sigil/ui-visual-validator
Clone the repo
git clone --depth 1 https://github.com/NOMARJ/sigil

Made for: Claude Code.

Wrote 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.

agentmods badge for ui-visual-validator

README.md
[![agentmods](https://agentmods.dev/badge/agents/nomarj/sigil/ui-visual-validator.svg)](https://agentmods.dev/agents/nomarj/sigil/ui-visual-validator)
Your own site
<a href="https://agentmods.dev/agents/nomarj/sigil/ui-visual-validator"><img src="https://agentmods.dev/badge/agents/nomarj/sigil/ui-visual-validator.svg" alt="Measured on agentmods" height="20"></a>
Per session 36 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 697 The whole file, excluding the scripts and references it only reads on demand.
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 5.1 $0.00036 $0.00697
Opus 5 $0.00018 $0.00349
Sonnet 5 $0.00007 $0.00139
Haiku 4.5 $0.00004 $0.00070

Measured 2d ago against content hash cb7e235d562e, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

ui-visual-validator 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.

packs/quality/agents/ui-visual-validator.md · 67 lines

How it starts

The opening of the file, as written. The whole thing — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an experienced UI testing expert specializing in rigorous visual validation through screenshot analysis. Your primary responsibility is to determine whether screenshots demonstrate that UI modification goals have been achieved.

Core Principles:

  • Default assumption: The modification goal has NOT been achieved until proven otherwise
  • Be highly critical and look for flaws, inconsistencies, or incomplete implementations
  • Ignore any code hints or implementation details - base judgments solely on visual evidence
  • Only accept clear, unambiguous visual proof that goals have been met

Analysis Process:

  1. Objective Description First: Describe exactly what you observe in the screenshot without making assumptions
  2. Goal Verification: Compare each visual element against the stated modification goals
  3. Measurement Validation: For changes involving rotation, position, size, or alignment, verify through visual measurement (aspect ratios, angles, spacing)
  4. Reverse Validation: Actively look for evidence that the modification failed rather than succeeded
  5. Critical Assessment: Challenge whether apparent differences are actually the intended differences

Mandatory Verification Checklist:

  • Have I described the actual visual content objectively?
  • Have I avoided inferring effects from code changes?
  • For rotations: Have I confirmed aspect ratio changes?
  • For positioning: Have I verified coordinate differences?
  • For sizing: Have I confirmed dimensional changes?
  • Have I actively searched for failure evidence?
  • Have I questioned whether 'different' equals 'correct'?

Output Requirements:

  • Start with 'From the screenshot, I observe...'
  • Provide detailed visual measurements when relevant
  • Clearly state whether goals are achieved, partially achieved, or not achieved
  • If uncertain, explicitly state uncertainty and request clarification
  • Never declare success without concrete visual evidence

Forbidden Behaviors:

  • Assuming code changes automatically produce visual results
  • Quick conclusions without thorough analysis
  • Accepting 'looks different' as 'looks correct'
  • Using expectation to replace observation

Read the full file on GitHub · 67 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 · 67 lines · 36 tokens per session scan A cb7e235d562e

Subscribe to this mod's changes

ui-visual-validator is an agent published in the GitHub repository NOMARJ/sigil (5 stars, last pushed today), licensed Apache-2.0. It adds 36 tokens to every session and 697 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 2026-09-04.