Borrowing it
Nothing to install: this file belongs to guillermoscript/lms-front. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/guillermoscript/lms-front/master/.agents/skills/ui-evidence/SKILL.mdgit clone --depth 1 https://github.com/guillermoscript/lms-frontWrote 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.
[](https://agentmods.dev/skills/guillermoscript/lms-front/ui-evidence)<a href="https://agentmods.dev/skills/guillermoscript/lms-front/ui-evidence"><img src="https://agentmods.dev/badge/skills/guillermoscript/lms-front/ui-evidence/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.
<a href="https://agentmods.dev/skills/guillermoscript/lms-front/ui-evidence"><img src="https://agentmods.dev/badge/skills/guillermoscript/lms-front/ui-evidence.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00121 | $0.00898 |
| Opus 5.5 | $0.00048 | $0.00359 |
| Sonnet 5.5 | $0.00024 | $0.00180 |
| Haiku 4.5 | $0.00012 | $0.00090 |
Grade A, and why
ui-evidence 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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
UI Evidence — before/after screenshots and a verification GIF
Reviewers weigh a before/after pair far more than a description of what changed, and the GIF comment is what tells them "this needs your visual sign-off". For any user-visible change this evidence is mandatory, not decorative.
Chrome is a single shared resource — never delegate it. All browser steps here (screenshots, GIF recording, uploads) run in the main agent via the claude-in-chrome tools; parallel subagents fighting over one browser serialize badly and corrupt each other's recordings. If several tasks need Chrome, queue their browser steps and do them one at a time at the end.
1. Before-captures — do this BEFORE touching the code
The only moment the "before" state exists is before implementation — once the change lands, it's gone. As soon as it's known that a change touches UI:
- Start (or find) the locally running app — check for a
runskill or repo docs for the start command and local URL/basePath conventions. - Open the affected screen(s) in Chrome and screenshot each
(
mcp__claude-in-chrome__computer, screenshot action), saved asissue-<N>-before-<screen>.png(orbefore-<screen>.pngif there's no issue number). - If the local app isn't running and starting it is disruptive right now, note that "before" captures were skipped and why — don't silently drop them.
2. Verification GIF + after-shots — after implementing
- Drive the changed flow end-to-end in the running app (the
verifyskill covers how to exercise it; this skill covers how to record it). - Record the drive with
mcp__claude-in-chrome__gif_creator: start capture before the first action, capture extra frames before and after each step so playback is smooth, and name the file after the work (e.g.issue-<N>-verification.gif). Save it somewhere durable and note the path. - Take "after" screenshots of the same screens captured in step 1
(
issue-<N>-after-<screen>.png) so the PR can show matched before/after pairs — same screen, same viewport, ideally same scroll position.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago First seen · 71 lines · 121 tokens per session scan A d67ad38a780b
ui-evidence is a skill published in the GitHub repository guillermoscript/lms-front (40 stars, last pushed today), licensed MIT. It adds 121 tokens to every session and 898 once invoked, about $0.0005 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 2026-08-30.
Other skills, from other repositories
chrome-cdp
Drive a headless Chrome over the Chrome DevTools Protocol (CDP) for browser QA — navigate, click, fill forms, read the DOM/accessibility tree, screenshot, and assert. Use whenever a task requires loading a web page and interacting with it like a user. Chrome is launched by a bash step (recipe below); this skill…
add-landing-template
Add a new org landing-page template to the ClassroomIO monorepo. Use when the user asks to "create/add a new landing template", "add a new template like X to the org landing pages", "build a new theme for the org landing page", or hands over a design reference (image, URL, prototype) for a new landing visual style.
prototype
Build multiple genuinely different versions of a UI piece you describe, rendered behind a visual picker so you can flip through them live and promote the one that feels right. Only runs when explicitly invoked; it does not trigger on its own.
ship-feature
End-to-end feature delivery — implement a feature, verify builds/tests pass, open a fully-filled draft PR, then automatically push every feedback round and review iteration to the same PR. Use when the user asks to implement or build a feature, fix-and-ship a change, or iterate on an existing feature PR.
write-prd
Write a product PRD for ClassroomIO the house way — clarify all features with the user first, research the codebase, write the PRD under prd/ /, then build clickable HTML prototypes under prototypes/ / that become the UX source of truth. Use when the user asks to "write a PRD", "spec a feature", "plan a new…
classroomio-design
ClassroomIO's design system. Use for any ClassroomIO UI, marketing page, launch image, social post, deck or thumbnail, and before choosing a colour, type size, radius or component. Covers tokens, app and marketing components, voice, and brand-asset rules.