A visual regression testing tool for comparing current screenshots with approved baseline screenshots. Visual regression testing finds unintended changes in a page's appearance.

It is for checking visual test reports, viewing detailed differences, and approving changed screenshots as new baselines only when a person explicitly requests it.

It summarizes changed, passed, new, and missing screenshots and separates possible layout, style, environment, and pixel differences for review.

testivai — Skill for Claude Code by testivai/testivai-oss

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.

Any agent

npx skills add testivai/testivai-oss --skill testivai

Clone the repo

git clone --depth 1 https://github.com/testivai/testivai-oss

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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/testivai/testivai-oss/testivai/github.svg)](https://agentmods.dev/skills/testivai/testivai-oss/testivai)

Your own site

<a href="https://agentmods.dev/skills/testivai/testivai-oss/testivai"><img src="https://agentmods.dev/badge/skills/testivai/testivai-oss/testivai/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/testivai/testivai-oss/testivai"><img src="https://agentmods.dev/badge/skills/testivai/testivai-oss/testivai.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 39 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,078 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.

Model	Per session	Once invoked
Fable 5.1	$0.00039	$0.01078
Opus 5.5	$0.00016	$0.00431
Sonnet 5.5	$0.00008	$0.00216
Haiku 4.5	$0.00004	$0.00108

Measured 3d ago against content hash df95b240be39, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

testivai 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 3d 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.

docs/claude-code/testivai/SKILL.md · 89 lines

How it starts

The opening of the file, as written. The whole thing — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/testivai

The person typed: /testivai $ARGUMENTS

Do only what that subcommand asks. Baselines are approved only because a person typed /testivai approve: never approve anything else, and never approve on your own in a later turn.

status (also when nothing was typed)

Read visual-report/results.json. If there is none, say there is no TestivAI report yet and that the visual tests write it.

Reply with the summary line (changed, passed, new and missing counts), then one line per changed snapshot: its name, diffPercent, and what the fields say, first match wins:

captureHealth.unhealthy set: unhealthy capture, with its reasons
sizeMismatch: the page size changed (baseline → current)
dom.changed: the DOM changed (dom.summary counts)
dom.styleCheck is "mismatch": a style-only change (dom.styleChanges)
environment.renderingDelta not empty: the environment changed since the baseline (those fields)
otherwise: pixels differ while the DOM and styles match

Whatever matched first, also name the elements in dom.styleChanges when dom.styleCheck is "mismatch" (a restyled element can sit next to a DOM change), and a pageShift (content moved down or up by its dy).

Then the missingBaselines. Never call a diff noise or harmless from these fields alone; the HTML report (/testivai report) gives the complete reading. Do not approve anything.

report [folder]

If the /testivai-report command is available (the TestivAI Claude Code plugin), tell the person to run /testivai-report [folder] and stop.

Otherwise show the report with its images inside the page: index.html loads its screenshots from images/, which a side panel or file viewer may not show. When the folder is the default visual-report and .testivai/temp/ has captures, run npx testivai report --share: it compares those captures again (the same comparison the tests ran) and writes visual-report/share.html, one file with every image inlined. Show that file if you can show files here, and give its path. If .testivai/config.json sets shareUploadCommand, never run it with --share: that would also upload the report. In that case, or with no captures in .testivai/temp/, or for another folder, give the path of <folder>/index.html and say its screenshots load from images/ next to it.

Read the full file on GitHub · 89 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.

3d ago Changed · +14 lines df95b240be39
4d ago First seen · 75 lines · 39 tokens per session scan A 57ae96bdb9e9

Subscribe to this mod's changes

testivai is a skill published in the GitHub repository testivai/testivai-oss (7 stars, last pushed today), licensed MIT. It adds 39 tokens to every session and 1,078 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-11-04.

E2E testing Testing ci github-actions playwright qa regression-testing screenshot-testing

testivai next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
testivaitestivai/testivai-oss	Skill	7	today	MIT	A	39	Claude Code
e2e-testingericrisco/rsc-harness	Skill	555	4d ago	MIT	A	79	Claude Code, Codex
Flaky Test DoctorPramodDutta/qaskills	Skill	245	6d ago	MIT	A	54	Claude Code, Codex
Checkly Monitoring as CodePramodDutta/qaskills	Skill	245	6d ago	MIT	A	35	Claude Code, Codex
CI Test Sharding ParallelizationPramodDutta/qaskills	Skill	245	6d ago	MIT	A	34	Claude Code, Codex
qawolf-cliqawolf/cli	Skill	3.5k	2d ago	Apache-2.0	C	112	Claude Code, Codex
test-reviewakovalion/paranoid-qa	Skill	16	6d ago	MIT	A	137	Claude Code