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.
npx skills add josephbartlett/windows-appshot-codex --skill windows-appshotgit clone --depth 1 https://github.com/josephbartlett/windows-appshot-codexWrote 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/josephbartlett/windows-appshot-codex/windows-appshot)<a href="https://agentmods.dev/skills/josephbartlett/windows-appshot-codex/windows-appshot"><img src="https://agentmods.dev/badge/skills/josephbartlett/windows-appshot-codex/windows-appshot.svg" alt="Measured on agentmods" 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 | $0.00050 | $0.00829 |
| Opus 5 | $0.00025 | $0.00415 |
| Sonnet 5 | $0.00010 | $0.00166 |
| Haiku 4.5 | $0.00005 | $0.00083 |
Grade A, and why
windows-appshot 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 7d 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 — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Windows Appshot
Use this skill when the user invokes the Windows Appshot plugin or when working with Windows appshot bundles generated by it.
Capture
Users normally invoke this as a Codex plugin command:
$windows-appshot create a Windows appshot
$windows-appshot capture Edge Gmail tab
Resolve <plugin-root> as two directories above this SKILL.md file:
skills/windows-appshot/SKILL.md -> plugin root -> scripts/.
When the user asks to create a new appshot without a target query, run:
& "<plugin-root>\scripts\New-Appshot.ps1"
The user has a short delay to focus the target app window. For hotkey capture, use:
& "<plugin-root>\scripts\Start-AppshotHotkey.ps1"
If the user includes a target query, pass it positionally:
& "<plugin-root>\scripts\New-Appshot.ps1" Edge Gmail tab
Query mode searches visible windows by process, title, and class name. It also inspects browser tab names from Edge, Chrome, Firefox, and Brave when available. Let the script ask for confirmation when matches are ambiguous or when a browser tab would be activated. Use -NoWindowConfirmation only for tests or trusted automation where the target is intentionally preselected.
Query captures verify the selected window is foreground before screenshot capture. Browser tab captures also verify selected-tab state through UI Automation when available or a strong active-title match.
For non-interactive Codex execution, prefer listing matches first and then selecting an explicit index:
& "<plugin-root>\scripts\New-Appshot.ps1" Edge Gmail tab -ListWindows
& "<plugin-root>\scripts\New-Appshot.ps1" Edge Gmail tab -TargetIndex 1 -NoWindowConfirmation
For regular window matches, -TargetIndex is an explicit selection after listing. For browser tab matches, it still requires -NoWindowConfirmation because tab activation can change app state.
Do not edit Codex session files directly. Use supported Codex inputs such as --image, codex exec resume --last --image, or explicit file references to the generated bundle.
The default command target is NewThread; ResumeLastExec must be explicit because it can send captured context to the wrong latest exec session.
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.
- 7d ago First seen · 74 lines · 50 tokens per session scan A 301f8a3ddf13
windows-appshot is a skill published in the GitHub repository josephbartlett/windows-appshot-codex (1 stars, last pushed 3mo ago), licensed MIT. It adds 50 tokens to every session and 829 once invoked, about $0.0003 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
design
Set visual and interaction direction for any UI surface (web, React, TUI, CLI, desktop, Qt, design-system tokens) before any UI code. Direction-first: generates 3-4 distinct directions via verbalized sampling, picks one via per-axis single-select, then derives palette, typography, spacing, motion budget. Loads when…
compose-creative-ui
A workflow for combining visual interface components and effects in an existing frontend while preserving its current design system and product identity. It covers component sources such as shadcn/ui, Magic UI, Radix, React Bits, and related tools.
implement-ui-interaction
A guide for implementing and testing frontend controls such as dialogs, menus, popovers, carousels, drag-and-drop, swiping, pinching, and other pointer, touch, keyboard, or focus interactions.
implement-ui-motion
A workflow for implementing motion and micro-interactions in an existing frontend. It covers CSS, the Web Animations API, View Transition API, Anime.js, Motion, and GSAP, including reduced-motion support.
uiux-advisor
A Korean-language UI and UX research guide for explaining concepts, comparing options, and writing product or interface specifications. It covers accessibility, content, interaction states, responsive design, ethics, testing, and performance.
uiux-auditor
A method for reviewing a user interface—the screens and interactions people use in a product—using screenshots, websites, prototypes, code, or described user flows. It checks observable problems and separates confirmed facts from assumptions and untested behaviour.