Borrowing it
Nothing to install: this file belongs to scrappy-builds/deskmail-ai. 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/scrappy-builds/deskmail-ai/main/.claude/skills/verify/SKILL.mdgit clone --depth 1 https://github.com/scrappy-builds/deskmail-aiWrote 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/scrappy-builds/deskmail-ai/verify)<a href="https://agentmods.dev/skills/scrappy-builds/deskmail-ai/verify"><img src="https://agentmods.dev/badge/skills/scrappy-builds/deskmail-ai/verify/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/scrappy-builds/deskmail-ai/verify"><img src="https://agentmods.dev/badge/skills/scrappy-builds/deskmail-ai/verify.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00024 | $0.00530 |
| Opus 5 | $0.00012 | $0.00265 |
| Sonnet 5 | $0.00005 | $0.00106 |
| Haiku 4.5 | $0.00002 | $0.00053 |
Grade A, and why
verify 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 10d 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.
What it actually says
Verifying DeskMail AI changes
Build + launch
npm run build # electron-vite build → out/
Drive the built app with Playwright's Electron support (already a dev dependency). From a Node .mjs script:
import { createRequire } from 'node:module'
const { _electron: electron } = createRequire(join(process.cwd(), 'package.json'))('playwright')
const app = await electron.launch({
args: [join(process.cwd(), 'out', 'main', 'index.js')],
env: { ...process.env, DESKMAIL_USER_DATA: mkdtempSync(join(tmpdir(), 'deskmail-verify-')) }
})
const win = await app.firstWindow()
DESKMAIL_USER_DATA points the app at a throwaway data dir — always set it so runs
don't touch real mail. Relaunch with the same dir to test persistence.
Driving gotchas
- Title-bar menus (File/Edit/View/…) are
<span>s, not buttons — usegetByText('File', { exact: true }), then menu items by text ('Settings…'). - Settings modal: open via File → Settings…; close by clicking the backdrop
(e.g.
win.mouse.click(8, 400)), the ✕ has no accessible name. - Buttons named
Delete/Editexist in the command bar behind modals too — never use baregetByRole('button', { name: 'Delete' }).nth(n); scope to the card/row (xpath//span[text()="…"]/parent::div//button[…]works). window.confirmdialogs: registerwin.on('dialog', d => d.accept())and logd.message()— the message tells you which item the click really hit.- The renderer bridge is available for state checks:
win.evaluate(() => window.deskmail.getSettings()). - Theme state lives on the root element:
data-themeattribute + inline--tokenstyle vars.
Worth driving
Theme/appearance: command-bar "Colour scheme" menu; Settings → Appearance (cards, New theme editor, live preview, Cancel-restores, delete-confirm). Native file dialogs (import/export/backup) can't be driven — verify around them.
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.
- 10d ago First seen · 51 lines · 24 tokens per session scan A ce99663b4d29
verify is a skill published in the GitHub repository scrappy-builds/deskmail-ai (1 stars, last pushed 29d ago), licensed Apache-2.0. It adds 24 tokens to every session and 530 once invoked, about $0.0001 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 2026-08-31.
Other skills, from other repositories
performance-engineer
Becomes a senior performance engineer who identifies bottlenecks, designs optimization strategies, and conducts load testing using systematic profiling and benchmarking methodology. Use when the user needs performance analysis, load testing, bottleneck identification, latency optimization, or capacity planning. Do NOT…
e2e-test-engineer
Use when the user asks to create, improve, review, or repair end-to-end tests for a meaningful user workflow across the UI, runtime, backend, or installed application. Validate the real workflow at the highest useful boundary while keeping environments, accounts, and test data controlled.
visual-regression-tester
Use when the user asks to detect, review, prevent, or update visual regressions in a web, desktop, or mobile interface. Capture comparable rendered states, control viewport and data variance, inspect image differences, and distinguish intended design changes from rendering defects.
playwright
Use when the task requires automating a real browser from the terminal (navigation, form filling, snapshots, screenshots, data extraction, UI-flow debugging) via playwright-cli or the bundled wrapper script.
develop-web-game
Use when Codex is building or iterating on a web game (HTML/JS) and needs a reliable development + testing loop: implement small changes, run a Playwright-based test script with short input bursts and intentional pauses, inspect screenshots/text, and review console errors with rendergametotext.
browser-qa
Use when validating or debugging a workflow in the embedded browser and you need a reproducible, evidence-first loop.