visual-verification
v1.1.1Makes the agent prove an output works before calling it done: scope the check from the user's words, verify on the real surface and environment (never tests, CI, health checks or localhost alone), screenshot it, use it (click, submit, run, open every page), read the evidence critically, and report each item as ✓/✗/?, in a strict tool order (real Chrome, in-app browser, headless Playwright, computer use), with a tool-traps guide, a rationalization table and red flags.
By Tomer HayunLicense: CC-BY-ND-4.00 GitHub starsUpdated 2 hours ago
Directory evidence
- Runtimes
- Claude Code
- Parsed components
- 1 skill or MCP entry
- Source updated
- Oct 3, 2026
- Manifest status
- Canonical path parsed
The directory validates manifest shape and source location. It does not execute the plugin or provide a security endorsement. Review the indexing methodology →
Install visual-verification for Claude Code
claude plugin marketplace add IchenDEV/agent-plugin-mkt
claude plugin marketplace update agent-plugin-marketplace
claude plugin install visual-verification@agent-plugin-marketplacePaste and run these commands in a terminal with Claude Code. They add and refresh the PluginsMP catalog, then install this plugin.
The installer fetches third-party code from the source repository shown on this page. This directory validates manifest structure and source location, but does not perform a security audit; review the manifest, components, and source before installing.
Get the source manually
git clone https://github.com/tomerhayundev/skillsClone the source repository, then follow its setup instructions to add the plugin to a compatible client. The plugin root is plugins/visual-verification/.
Plugin files
├── .claude-plugin/plugin.json└── skills/visual-verification/SKILL.md
Included Skills1
Use when about to say work is done, fixed, working, verified, live, or ready, or to hand off or approve an output: after merging, deploying, or pushing to staging or production; after publishing or republishing an artifact or design; when the user asks "is it fixed / working?" or to confirm, test, or visually verify something; when a stop hook or /goal asks if the task is complete. Especially when the evidence is code, tests/CI/E2E, a health check or status code, a harness or dev route, localhost, JS measurements, or a screenshot of a feature you never used. Covers UI, pages, interactive flows, logins/OAuth, exports and PDFs, designs, generated media, documents, CLI/API output, prose and translations, and data.
Plugin manifests1
{
"name": "visual-verification",
"version": "1.1.1",
"description": "Makes the agent prove an output works before calling it done: scope the check from the user's words, verify on the real surface and environment (never tests, CI, health checks or localhost alone), screenshot it, use it (click, submit, run, open every page), read the evidence critically, and report each item as ✓/✗/?, in a strict tool order (real Chrome, in-app browser, headless Playwright, computer use), with a tool-traps guide, a rationalization table and red flags.",
"author": {
"name": "Tomer Hayun",
"url": "https://github.com/tomerhayundev"
},
"homepage": "https://github.com/tomerhayundev/skills/tree/main/plugins/visual-verification",
"license": "CC-BY-ND-4.0",
"keywords": [
"verification",
"testing",
"qa",
"browser",
"screenshots"
]
}For maintainers
If you maintain this plugin, link to this source-backed listing from your README so users can review its manifest and indexed components.
[visual-verification on Agent Plugins Marketplace](https://pluginsmp.com/plugins/visual-verification)