<!-- mobian-agent-page publisher="dailydev" canonical="https://daily.dev/posts/beyond-autocomplete-getting-the-most-from-github-copilot-for-net-8jm5hyma6" -->

---
title: Beyond Autocomplete: Getting the Most from GitHub...
description: A Microsoft .NET product manager demos fixing a duplicate purchase-order bug across Visual Studio, VS Code, the Copilot app, and Copilot CLI using GitHub...
canonical: https://daily.dev/posts/beyond-autocomplete-getting-the-most-from-github-copilot-for-net-8jm5hyma6
twitter:card: summary_large_image
twitter:site: @dailydotdev
og:type: website
og:site_name: daily.dev
og:title: Beyond Autocomplete: Getting the Most from GitHub Copilot for .NET | daily.dev
og:description: A Microsoft .NET product manager demos fixing a duplicate purchase-order bug across Visual Studio, VS Code, the Copilot app, and Copilot CLI using GitHub...
og:url: https://daily.dev/posts/beyond-autocomplete-getting-the-most-from-github-copilot-for-net-8jm5hyma6
og:image: https://api.daily.dev/og/posts/8jM5HymA6.png
og:image:alt: Beyond Autocomplete: Getting the Most from GitHub Copilot for .NET
og:image:width: 1200
og:image:height: 630
og:locale: en
---

> ## Documentation Index
> Fetch the complete documentation index at: https://daily.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Beyond Autocomplete: Getting the Most from GitHub Copilot for .NET

**[.NET Blog](https://daily.dev/sources/dotnet)** · 23 min read · 0 upvotes · 0 comments

## Summary

A Microsoft .NET product manager demos fixing a duplicate purchase-order bug across Visual Studio, VS Code, the Copilot app, and Copilot CLI using GitHub Copilot's agentic features. Copilot auto-detects installed .NET skills like the WinForms expert and web API skills based on the solution. A new skill-value dashboard at github.com/dotnet/skills shows token usage and pass-rate improvements per model. NuGet packages can now ship their own Copilot skills via a 'dotnet package skills' command. Roslyn's new daemon mode cuts memory usage about 50% when running multiple agents simultaneously, and a 170-scenario .NET-specific benchmark catches regressions like a tool that spiked token usage before release.

## Full article

daily.dev links to this article rather than hosting it. Read it at the original source: <https://www.youtube.com/watch?v=R4ZgrOcazT4>

## Questions this post answers

### How does Roslyn's daemon mode reduce memory usage when running multiple Copilot agents at once?

Roslyn's new daemon mode shares resources across multiple agents instead of spinning up separate processes for each, cutting memory usage by about 50% in multi-agent scenarios. It is enabled through the .NET skills plugin configuration, which sets up agents to use LSP with Roslyn in daemon mode for code search, letting systems handle three or four simultaneous agents without slowing down.

_Running several coding agents at once? daily.dev surfaces updates like Roslyn's daemon mode before memory issues hit._

### Can NuGet packages ship their own Copilot agent skills?

Yes, a new feature lets package maintainers bundle Copilot skills directly inside their NuGet package. When a developer runs 'dotnet restore', those skills download automatically to their machine, and a 'dotnet package skills' command copies them into the folder agents use to recognize and apply them. This was announced in a blog post by Richlander shortly before the demo.

_Tracking new .NET tooling capabilities like packaged skills is easier with daily.dev in your workflow._

### How does Microsoft test GitHub Copilot's reliability specifically for C# and .NET code?

The .NET team built a custom benchmark of roughly 170 scenarios run regularly against top and frontier models, since mainstream agentic benchmarks contain no .NET or C# code. Visual Studio, VS Code, and the Copilot SDK teams all use it to catch regressions; it once caught a tool that boosted token usage past an acceptable threshold before release, and it led to a new C# refactoring skill after refactoring scenarios on large codebases kept failing.

_Choosing an AI coding assistant for a C# codebase gets easier when you track benchmark results like these on daily.dev._

## Similar posts on daily.dev

- [Extend your coding agent with .NET Skills](https://daily.dev/posts/extend-your-coding-agent-with-net-skills-akfma6i5a) · .NET Blog · 89 upvotes · 3 comments
- [Doing More with GitHub Copilot as a .NET Developer](https://daily.dev/posts/doing-more-with-github-copilot-as-a-net-developer-dektd3qud) · ASP.NET Blog · 34 upvotes · 4 comments
- [GitHub Copilot CLI for .NET Developers: Real Workflows on a C\# Codebase](https://daily.dev/posts/github-copilot-cli-for-net-developers-real-workflows-on-a-c-codebase-eyicrg3rl) · We Are .NET · 4 upvotes · 0 comments
- [Microsoft Adds Custom Copilot Agents for .NET Developers with C\# and WinForms Experts](https://daily.dev/posts/microsoft-adds-custom-copilot-agents-for-net-developers-with-c-and-winforms-experts-ckcufr6rv) · InfoQ · 48 upvotes · 1 comments

---

Tags: [#github](https://daily.dev/tags/github), [#.net](https://daily.dev/tags/.net), [#ai-coding](https://daily.dev/tags/ai-coding), [#vscode](https://daily.dev/tags/vscode), [#visual-studio](https://daily.dev/tags/visual-studio)

[View this post on daily.dev](https://daily.dev/posts/beyond-autocomplete-getting-the-most-from-github-copilot-for-net-8jm5hyma6)

```json
{"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]}
{"@context":"https://schema.org","@type":"TechArticle","headline":"Beyond Autocomplete: Getting the Most from GitHub Copilot for .NET","url":"https://daily.dev/posts/beyond-autocomplete-getting-the-most-from-github-copilot-for-net-8jm5hyma6","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/beyond-autocomplete-getting-the-most-from-github-copilot-for-net-8jm5hyma6"},"datePublished":"2026-11-12T01:01:58.383Z","dateModified":"2026-11-12T01:03:28.281Z","description":"A Microsoft .NET product manager demos fixing a duplicate purchase-order bug across Visual Studio, VS Code, the Copilot app, and Copilot CLI using GitHub...","image":"https://i.ytimg.com/vi/R4ZgrOcazT4/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/R4ZgrOcazT4/sddefault.jpg","isAccessibleForFree":true,"articleSection":".NET Blog","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":".NET Blog","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/f23eb59371bd4f07aeef64c4eef73266","url":"https://daily.dev/sources/dotnet"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/beyond-autocomplete-getting-the-most-from-github-copilot-for-net-8jm5hyma6","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"github,.net,ai-coding,vscode,visual-studio","timeRequired":"PT23M","video":{"@type":"VideoObject","name":"Beyond Autocomplete: Getting the Most from GitHub Copilot for .NET","description":"A Microsoft .NET product manager demos fixing a duplicate purchase-order bug across Visual Studio, VS Code, the Copilot app, and Copilot CLI using GitHub...","thumbnailUrl":"https://i.ytimg.com/vi/R4ZgrOcazT4/sddefault.jpg","uploadDate":"2026-11-12T01:01:58.383Z","duration":"PT23M","url":"https://api.daily.dev/r/8jM5HymA6","embedUrl":"https://www.youtube.com/embed/R4ZgrOcazT4"}}
{"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":".NET Blog","item":"https://daily.dev/sources/dotnet"},{"@type":"ListItem","position":3,"name":"Beyond Autocomplete: Getting the Most from GitHub Copilot for .NET"}]}
{"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/beyond-autocomplete-getting-the-most-from-github-copilot-for-net-8jm5hyma6#faq","mainEntity":[{"@type":"Question","name":"How does Roslyn's daemon mode reduce memory usage when running multiple Copilot agents at once?","acceptedAnswer":{"@type":"Answer","text":"Roslyn's new daemon mode shares resources across multiple agents instead of spinning up separate processes for each, cutting memory usage by about 50% in multi-agent scenarios. It is enabled through the .NET skills plugin configuration, which sets up agents to use LSP with Roslyn in daemon mode for code search, letting systems handle three or four simultaneous agents without slowing down. Running several coding agents at once? daily.dev surfaces updates like Roslyn's daemon mode before memory issues hit."}},{"@type":"Question","name":"Can NuGet packages ship their own Copilot agent skills?","acceptedAnswer":{"@type":"Answer","text":"Yes, a new feature lets package maintainers bundle Copilot skills directly inside their NuGet package. When a developer runs 'dotnet restore', those skills download automatically to their machine, and a 'dotnet package skills' command copies them into the folder agents use to recognize and apply them. This was announced in a blog post by Richlander shortly before the demo. Tracking new .NET tooling capabilities like packaged skills is easier with daily.dev in your workflow."}},{"@type":"Question","name":"How does Microsoft test GitHub Copilot's reliability specifically for C# and .NET code?","acceptedAnswer":{"@type":"Answer","text":"The .NET team built a custom benchmark of roughly 170 scenarios run regularly against top and frontier models, since mainstream agentic benchmarks contain no .NET or C# code. Visual Studio, VS Code, and the Copilot SDK teams all use it to catch regressions; it once caught a tool that boosted token usage past an acceptable threshold before release, and it led to a new C# refactoring skill after refactoring scenarios on large codebases kept failing. Choosing an AI coding assistant for a C# codebase gets easier when you track benchmark results like these on daily.dev."}}]}
```

