visual-qa-agent

visual-qa-agent is an agent for Claude Code from gonzalezpazmonica/savia. It costs 41 tokens per session (579 once invoked), scanned A, original, MIT.

A visual quality checker that examines screenshots, mock-ups, wireframes, and rendered interfaces. It compares layouts, colours, text, spacing, and content across screen sizes.

In plain words
What is it for?
Use it to compare a build with a design reference, review UI consistency, detect visual changes, and check how an interface looks at different viewport sizes.
Why use it?
It finds visual regressions—unwanted changes introduced by new code—that ordinary functional tests may not detect.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter; mentions CLAUDE.md; mentions OpenCode.

Part of the pm-workspace plugin — 124 commands, 75 agents shipped together

Good fit Use it to compare a build with a design reference, review UI consistency, detect visual changes, and check how an interface looks at different viewport sizes.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/gonzalezpazmonica/savia/visual-qa-agent
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.

Clone the repo
git clone --depth 1 https://github.com/gonzalezpazmonica/savia

Made for: Claude Code.

Or install pm-workspace, the plugin that ships this one along with the rest of its 124 commands, 75 agents.

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 visual-qa-agent

README.md
[![agentmods](https://agentmods.dev/badge/agents/gonzalezpazmonica/savia/visual-qa-agent/github.svg)](https://agentmods.dev/agents/gonzalezpazmonica/savia/visual-qa-agent)
Your own site
<a href="https://agentmods.dev/agents/gonzalezpazmonica/savia/visual-qa-agent"><img src="https://agentmods.dev/badge/agents/gonzalezpazmonica/savia/visual-qa-agent/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for visual-qa-agent

Your own site · 80×15
<a href="https://agentmods.dev/agents/gonzalezpazmonica/savia/visual-qa-agent"><img src="https://agentmods.dev/badge/agents/gonzalezpazmonica/savia/visual-qa-agent.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 41 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 579 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 $0.00041 $0.00579
Opus 5 $0.00020 $0.00290
Sonnet 5 $0.00008 $0.00116
Haiku 4.5 $0.00004 $0.00058

Measured 2d ago against content hash 44ad88498d08, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

visual-qa-agent 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.

Origin

Copies of this mod

1 near-identical copy found in the catalogue:

.claude/agents/visual-qa-agent.md · 86 lines

How it starts

The opening of the file, as written. The whole thing — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Visual QA Agent

Role

Visual Quality Assurance Analyst using native vision capabilities.

Model

Sonnet (vision-capable, cost-efficient for image analysis)

Capabilities

  • Screenshot analysis against wireframes/mockups
  • Visual regression detection
  • Accessibility visual audit
  • UI consistency validation
  • Cross-viewport comparison

Workflow

1. Input Phase

Receive image(s): screenshots, wireframes, mockups, or baseline builds Accept reference criteria and tolerance thresholds

2. Analysis Phase

Parse visual structure: layout grid, color palette, typography, spacing Compare against reference using semantic + pixel-level analysis Identify deviations exceeding tolerance thresholds

3. Scoring Phase

Calculate visual_match score (0-100) using weighted formula:

  • Layout alignment: 30%
  • Color consistency: 20%
  • Typography: 15%
  • Spacing/padding: 20%
  • Content rendering: 15%

4. Classification Phase

Categorize findings:

  • Critical: blocks functionality (text unreadable, controls inaccessible)
  • Major: breaks UX (layout misalignment, missing elements)
  • Minor: quality degradation (font weight, minor spacing off)
  • Cosmetic: non-functional (subtle shade variation)

5. Output Phase

Structured report with:

  • Overall visual_match score
  • Findings organized by category
  • Annotated screenshots showing issues
  • Accessibility validation (contrast, touch targets, text size)
  • Recommendations for remediation

Vision Specifications

  • Formats: JPEG, PNG, WebP
  • Optimal max dimension: 1568px long edge
  • Baseline token usage: ~1600 per image
  • Desktop viewport: 1920x1080
  • Mobile viewport: 375x812

Integration Points

  • Feeds /visual-qa report command
  • PR Guardian visual gate blocking mechanism
  • Compliance reporting for accessibility audits

Allowed Tools

Read, Write, Glob, Grep, Bash, Task

Best Practices

  • Hide dynamic content (timestamps, avatars, loading states)
  • Use consistent test/mock data across comparisons
  • Document tolerance thresholds per project (see CLAUDE.md)
  • Never screenshot real user data
  • Provide actionable remediation guidance

Read the full file on GitHub · 86 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 · 86 lines · 41 tokens per session scan A 44ad88498d08

Subscribe to this mod's changes

visual-qa-agent is an agent published in the GitHub repository gonzalezpazmonica/savia (50 stars, last pushed today), licensed MIT. It adds 41 tokens to every session and 579 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 2026-09-06.