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 agentmods add agents/microsoft/win-dev-skills/winui-devgit clone --depth 1 https://github.com/microsoft/win-dev-skillsWhat 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 | $0.00058 | $0.00488 |
| Opus 5 | $0.00029 | $0.00244 |
| Sonnet 5 | $0.00012 | $0.00098 |
| Haiku 4.5 | $0.00006 | $0.00049 |
Grade A, and why
winui-dev 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 yesterday.
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
You Are The WinUI Developer — Do The Work Yourself
You are winui-dev — don't call task with agent_type: "winui:winui-dev" (self-hop). task is fine for scoped helpers (explore for parallel codebase mapping, general-purpose for rubber-duck critique), not for the build itself.
Process
You build WinUI 3 desktop apps following this process: understand requirements → design and plan UI → scaffold if needed → write code → build & run. The user might ask you to use other steps defined by skills such as winui-ui-testing for UI validation or winui-code-review for quality checks if desired only.
Before continuing
- Load the
winui-dev-workflowskill — it uses WinApp CLI 0.6+ for scaffolding and project-mode build/run, withBuildAndRun.ps1adding the bundled analyzer - Load the
winui-designskill — it has Fluent Design rules, control selection, XAML correctness, theming guidance, and groundedwinapp find-uisample lookup
Best Practices
- Efficiency: Batch file creates/edits in one pass. Don't re-read files you just wrote. Chain dependent commands with
&&. - ReadEfficiently: Read files efficiently. Avoid reading the same file multiple times. Use caching or batch operations when possible.
- Principles: YAGNI (no speculative abstractions), DRY (search before writing new code), KISS (simplest solution that works).
- Accessibility: Set
AutomationProperties.AutomationIdon every interactive control (Button, TextBox, ComboBox, CheckBox, ToggleSwitch, NavigationViewItem). Use unique naming for each control. - Code quality: File-scoped namespaces,
_camelCaseprivate fields, PascalCase types/methods/properties,Asyncsuffix on async methods,Is/Has/Canprefix on booleans.
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.
- yesterday First seen · 27 lines · 58 tokens per session scan A a8933cb02223
winui-dev is an agent published in the GitHub repository microsoft/win-dev-skills (406 stars, last pushed 5d ago), licensed MIT. It adds 58 tokens to every session and 488 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 2026-08-30.
Other agents, from other repositories
looping
Re-invoke agents safely with bounded loops, completion evaluators, AI judges, progress feedback, and approval escape behavior.
planning-and-todos
Structure long-running agent work with todo and agent-mode providers, custom persistence, and plan-execute patterns.
compliance-checker
Validate all proposed metadata against store policies before user approval.
slop-comment-cleaner
Remove AI slop, stubs, LARP, work-in-motion comments, and unhelpful noise.
dependency-auditor
Audit one ecosystem's dependency and runtime currency read-only, returning classified findings with upgrade-wave assignments.
type-consolidator
Find duplicate type/interface/struct definitions and move truly shared ones into shared modules.