Official agent skill

Agentic Labeler

by dotnet in dotnet/maui

Labels issues and pull requests in the dotnet/maui repository with area- and platform/ labels ONLY, based on technical content and platform-file conventions.

OfficialMITAuto-check passedDevelopment

Install Agentic Labeler

skills CLI
$ npx skills add dotnet/maui --skill agentic-labeler -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install dotnet/maui agentic-labeler --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/dotnet/maui.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.github/skills/agentic-labeler .claude/skills/agentic-labeler && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
agentic-labeler
GitHub stars
23k
Token cost
~3.5k tokens
SKILL.md length
1,652 words
Files
2
Skills in repo
27
Repo updated
First seen
Licence
MIT

At a glance

Labels issues and pull requests in the dotnet/maui repository with area- and platform/ labels ONLY, based on technical content and platform-file conventions.

  • Works in 2 steps: Exactly one area-* — derived from the… → One or more platform/* — derived from…
  • Tasks that involve Pull requests
  • SKILL.md covers 🚨 Scope: area-* and…, Label discovery and Labeling rules
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Agentic Labeler is an agent skill from dotnet/maui, published by the product's own GitHub organization. Labels issues and pull requests in the dotnet/maui repository with area- and platform/ labels ONLY, based on technical content and platform-file conventions. Used by the gh-aw agentic-labeler workflow and available for batch evaluation and interactive Copilot CLI usage.

Its SKILL.md is about 3.5k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files (for example `tests/eval.vally.yaml`).

It sits in Development, covering Pull requests. It works with .NET. The repository describes itself as: .NET MAUI is the .NET Multi-platform App UI, a framework for building native device applications spanning mobile, tablet, and desktop. The licence is MIT.

When your agent uses it

  • Tasks that involve Pull requests

Example prompts

  • “Use the agentic-labeler skill to label issues and pull requests in the dotnet/maui repository with area- and platform/ labels ONLY, based on…”
  • “/agentic-labeler”

Workflow steps

2 steps, taken from the first numbered list in SKILL.md.

  1. Exactly one area-* — derived from the subject matter (control name, area like layout / navigation / xaml / infrastructure / etc.). Choose…
  2. One or more platform/* — derived from changed-file platform conventions on PRs, or from explicit platform mentions on issues. Apply all…

What it can do on your machine

Read from SKILL.md and the folder at commit b926f05. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Agentic Labeler loads about 3.5k tokens when it runs. Until then it costs about 73 tokens; SKILL.md has 1,652 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~73
When it runs · the whole SKILL.md, loaded when a task matches
~3.5k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from dotnet/maui at commit b926f05, republished under its MIT licence (© dotnet). 1,652 words, ~3,457 tokens.

Download SKILL.mdSave it as .claude/skills/agentic-labeler/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
agentic-labeler
description
Labels issues and pull requests in the dotnet/maui repository with `area-*` and `platform/*` labels ONLY, based on technical content and platform-file conventions. Used by the gh-aw agentic-labeler workflow and available for batch evaluation and interactive Copilot CLI usage.
metadata.author
dotnet-maui
metadata.version
2.0

Agentic Labeler

Labeling rules for the dotnet/maui repository. These rules are the canonical source of truth for how issues and PRs should be labeled. They are consumed by the agentic-labeler gh-aw workflow and can also be used standalone for batch evaluation or interactive labeling.

🚨 Scope: area-* and platform/* ONLY

The labeler applies only two label families, and nothing else:

  1. Exactly one area-* — derived from the subject matter (control name, area like layout / navigation / xaml / infrastructure / etc.). Choose the single most specific match for the dominant subsystem; see the tie-breaking rules below.
  2. One or more platform/* — derived from changed-file platform conventions on PRs, or from explicit platform mentions on issues. Apply all that fit. Exception: platform/tizen is never applied by this labeler under any circumstance, even when Tizen files are touched or Tizen is named in an issue (see the platform-label rules below).

The labeler must NOT apply any other label, ever. Specifically, do not apply:

  • t/* (kind: t/bug, t/enhancement ☀️, t/docs 📝, t/breaking 💥, t/native-embedding, t/desktop, t/a11y, etc.) — the issue/PR author or other automation owns these.
  • i/* (indicators: i/regression, etc.) — set during triage based on investigation, not initial content.
  • s/* (status: s/needs-info, s/needs-repro, s/needs-verification, s/needs-attention, s/triaged, s/verified, s/no-repro, s/not-a-bug, s/duplicate 2️⃣, s/pr-needs-author-input, etc.) — managed by dotnet-policy-service[bot] and human triagers.
  • p/* (priority: p/0, p/1, p/2, p/3) — set by maintainers.
  • partner/* (e.g., partner/syncfusion) — set by partner-tracking automation.
  • perf/* (e.g., perf/memory-leak 💦) — set during perf investigation.
  • backport/*, regressed-in-*, version/* — set during triage / release management.
  • untriaged, :watch: Not Triaged — applied by repo automation on issue open.
  • Anything else that is not literally an area-* or platform/* label.

If the only labels that clearly apply are not area-* or platform/*, noop instead — see the noop section below.

If neither an area-* nor a platform/* label clearly applies, noop.

Label discovery

  • Fetch the current list of labels using the list_label MCP tool (provided by the labels toolset). Note the singular name — it is list_label, not list_labels.
  • Important pagination caveat: the list_label tool only returns the first ~100 labels (no pagination). This repo has ~440 labels, so many area-* and platform/* labels will be missing from the listing. If you have a strong candidate area-* or platform/* label name in mind that isn't in the listing, verify it exists with the get_label tool before adding it.
  • Do not create new labels — only labels that already exist in the repository will be accepted.

Labeling rules

area-* label (issues and PRs) — exactly one

Apply exactly one area-* label. Pick the single most specific match for the dominant subsystem:

  • Specific control mentioned → matching area-controls-<name> (e.g., CollectionView → area-controls-collectionview, Entry → area-controls-entry, Map / Maps → area-controls-map, Window → area-controls-window, WebView → area-controls-webview, HybridWebView → area-controls-hybridwebview). Always use the area-controls-<name> prefix — never invent shorter aliases (e.g., the Maps area is area-controls-map, not area-maps).
  • Layout, measure/arrange, sizing issues → area-layout.
  • Navigation, Shell routing, page navigation → area-navigation (or area-controls-shell when Shell-specific).
  • XAML parsing, markup extensions, XamlC, source generators → area-xaml.
  • Installation, workload availability, target-framework recognition, requirements, and platform support → area-setup.
  • Hot reload, debugging, editor experiences, build tasks, and MSBuild/tooling behavior → area-tooling.
  • Project templates → area-templates.
  • BlazorWebView / Blazor hybrid → area-blazor.
  • Essentials APIs (non-UI: connectivity, sensors, preferences, etc.) → area-essentials.
  • Drawing / Microsoft.Maui.Graphics → area-drawing.
  • Gestures (tap, pan, swipe, pinch) → area-gestures.
  • Lifecycle, hosting, app startup, DI → area-core-lifecycle / area-core-hosting.
  • Dispatcher / main thread / threading → area-core-dispatching.
  • Localization / RTL / culture → area-localization.
  • Docs only → area-docs.
  • Copilot CLI agents, agent skills, agentic workflows, and AI-assisted development → area-ai-agents. This covers .github/agents/**, .github/skills/**, gh-aw workflow sources and generated locks, and supporting review/test/fix orchestration when the AI-agent behavior is the primary subject.
  • CI, build pipelines, Maestro / dependency flow, branch mirroring, and generic GitHub Actions infrastructure → area-infrastructure. This covers:
    • [dnceng-bot] codeflow/branch-mirroring issues (the standard "Branch … can't be mirrored to Azdo" issues) → area-infrastructure (do not noop these — they have a clear area).
    • PRs touching only generic .github/workflows/*.yml, .github/scripts/, eng/pipelines/, eng/common/, or other CI infrastructure files → area-infrastructure (prefer this over area-tooling, which is for the dev-build/MSBuild/workload surface that ships to users).
    • AI-agent vs infrastructure tie-break: choose area-ai-agents when the change primarily alters an agent's behavior, prompt, skill, evaluation, or AI-assisted developer workflow. Command parsing and trigger/dispatch behavior for AI-assisted review count as agent behavior. Choose area-infrastructure when the change primarily alters generic CI execution, authentication, scheduling, or pipeline plumbing without changing the AI-assisted workflow's behavior, even if an agent consumes that infrastructure.
    • Mixed PRs (infra-primary + small product edits): if the PR is dominated by CI infrastructure changes but also has incidental edits to product code, still apply area-infrastructure (and omit any product area-*). If the product-code change is the focus and the infra change is incidental (e.g., a small workflow tweak that supports a feature), prefer the product area-* label and omit area-infrastructure.

Tie-breaking when multiple areas could apply — pick the single most specific:

  • Specific control beats generic area. area-controls-tabbedpage over area-navigation; area-controls-collectionview over area-layout; area-controls-shell over area-navigation.
  • Sub-area beats parent area. area-safearea over area-layout; area-core-dispatching over area-core-lifecycle.
  • Subject-matter focus beats incidental touch. If a PR fixes a CollectionView bug by adjusting layout code, the area is the control (area-controls-collectionview), not the layout system.
  • When genuinely tied, prefer the area that names the user-visible feature over the implementation-detail area.

If after applying these heuristics there is still no single best fit, noop rather than apply two area labels.

Show full SKILL.md (796 more words)Show less
platform/* labels

This is the most important behavior for PRs.

For pull requests, infer platform/* labels primarily from the changed files, using the rules below. Each rule maps a file pattern to one or more platform labels. Apply a platform/* label if any changed file matches that pattern. The path patterns intentionally target the established MAUI source-layout conventions — match the patterns in the table below (e.g., /Platform/<Name>/, /Platforms/<Name>/, /Handlers/*/<Name>/). Do not match on a bare top-level /Android/, /iOS/, /Windows/, or /MacCatalyst/ segment that is not part of one of the patterns in the table — bare segments occur in templates, docs, and unrelated tooling paths and are not platform-specific source code.

Note on iOS / MacCatalyst: file-extension patterns and directory patterns map differently because of MAUI's compilation conventions — they are split into separate rows below.

File pattern (changed in the PR)Label(s) to apply
*.android.cs, *.Android.cs, paths containing /Platform/Android/, /Platforms/Android/, /AndroidNative/, or handler subdirectories like /Handlers/*/Android/platform/android
*.ios.cs, *.iOS.cs (file-extension pattern — these compile for both iOS and MacCatalyst)platform/ios and platform/macos
Paths containing /Platform/iOS/, /Platforms/iOS/, or handler subdirectories like /Handlers/*/iOS/ (directory pattern — these compile only for the iOS TFM)platform/ios only
*.maccatalyst.cs, *.MacCatalyst.cs, paths containing /Platform/MacCatalyst/, /Platforms/MacCatalyst/, or handler subdirectories like /Handlers/*/MacCatalyst/platform/macos
*.windows.cs, *.Windows.cs, paths containing /Platform/Windows/, /Platforms/Windows/, or handler subdirectories like /Handlers/*/Windows/platform/windows

Notes:

  • If a PR touches only shared / cross-platform code (e.g., src/Core/src/*.cs without a platform suffix, or src/Controls/src/Core/), do not apply any platform/* label.
  • If a PR touches multiple platforms, apply each matching platform/* label.
  • .ios.cs files compile for both iOS and MacCatalyst (see split table rows above).
  • .maccatalyst.cs files do not compile for iOS — apply only platform/macos for those.
  • Tizen is excluded. *.tizen.cs files and /Platform/Tizen/ or /Platforms/Tizen/ paths exist in the source tree, but platform/tizen is never auto-applied. Treat Tizen files as if they had no platform suffix for labeling purposes: pick an area-* label normally based on the code's subject matter (e.g., TabbedPage.tizen.cs → area-controls-tabbedpage) and apply no platform/* label for the Tizen content. Only noop if the global noop rules below (e.g., no area-* clearly fits) apply on their own merits.

For issues, infer platform/* labels only for platforms the reporter explicitly identifies as affected: Android, iOS, macOS / Mac Catalyst, or Windows. This includes the issue-template's "Affected platforms" field, plus clear evidence in the title, body, or attached logs/stack traces. Do not guess.

  • If the reporter explicitly lists named affected platforms, apply one platform/* label per named platform — even when the list covers every supported platform (e.g., "iOS, Android, Windows, macOS" → apply all four).
  • Generic phrases like "all platforms", "every platform", "any platform", or "cross-platform" do not by themselves justify any platform/* label. If a generic phrase is accompanied by an explicit affected-platform list, the named list wins — apply each named platform's label (e.g., "all platforms (iOS, Android, Windows, macOS)" → apply all four).
  • Do not apply labels for platforms mentioned incidentally ("tested on iOS"), as "not reproduced" / "not affected", or as label requests ("please add platform/android").
  • Never apply platform/tizen — even when the reporter names Tizen, includes Tizen in an "Affected platforms" enumeration, or attaches Tizen logs. Tizen mentions are silently dropped from the platform set.
When to noop (no labels)

Some items should not be labeled. If any of the following apply, skip labeling entirely:

  • Automated inter-branch merge PRs — titles like [automated] Merge branch 'main' => 'net11.0' or similar bot-created merge PRs. These are infrastructure, not feature/bug work.
  • Dependency bump PRs that already have dependencies and area-infrastructure labels.
  • Items where no area-* or platform/* label clearly fits — when the content is too vague or ambiguous to determine area or platform with confidence, or when the only labels that would apply are outside the allowed area-* / platform/* scope.

⚠️ Do NOT noop [dnceng-bot] codeflow/branch-mirroring issues. Despite being bot-authored, they have a clear area (area-infrastructure) and should be labeled, not noop'd. The noop rule for automated PRs above is specifically about [automated] Merge branch … titles.

What NOT to do
  • Do not apply any label that is not literally area-* or platform/*. No t/*, i/*, s/*, p/*, partner/*, perf/*, backport/*, regressed-in-*, version/*, untriaged, :watch: Not Triaged, or anything else. See the "Scope" section at the top for the full prohibition.
  • Do not apply platform/tizen under any circumstance — Tizen is excluded from this labeler even when Tizen files are touched or Tizen is named in an issue body, title, or affected-platforms field.
  • Do not create new labels — apply only labels that already exist in the repository.
  • Do not add platform/* labels to PRs that don't touch platform-specific files.
  • Do not post a comment summarizing the labels — labels speak for themselves.
  • Do not close, lock, or otherwise modify the issue/PR beyond labeling.
  • Do not label automated merge PRs — these are infrastructure, not actionable items.
  • Be conservative; precision beats recall. Only apply area-* or platform/* labels that clearly fit.

© dotnet, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in .github/skills/agentic-labeler of dotnet/maui.

  • SKILL.md
  • tests/eval.vally.yaml

Open the folder on GitHubat commit b926f05

Compare with similar skills

Agentic Labeler next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Agentic Labeler compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Agentic Labeler this skilldotnet/maui23k—~3.5kAutomated safety check: PassMIT
.NET MAUI Code Reviewdotnet/efcore15k—~2kAutomated safety check: PassMIT
Code Reviewjonathanpeppers/dotnes780—~2.1kAutomated safety check: PassMIT
Open Pull RequestGremlinq/ExRam.Gremlinq187—~1.5kAutomated safety check: PassMIT
Gh Stackdotnet/macios2.9k3 repos~10kAutomated safety check: PassCustom licence
Pull RequestOpenCoreMMO/OpenCoreMMO481—~814Automated safety check: PassGPL-3.0

Similar skills

  • Official

    Deep code-only review of a pull request or candidate patch for correctness, safety and .NET MAUI conventions, judging the code before reading the PR description.

    15k GitHub stars~2k tokensUpdated today
    DevelopmentAuto-check passed
  • Code Review

    jonathanpeppers/dotnes

    Review dotnes pull requests against established repository rules.

    780 GitHub stars~2.1k tokensUpdated 14 days ago
    DevelopmentAuto-check passed
  • Open Pull Request

    Gremlinq/ExRam.Gremlinq

    A skill your agent uses when opening a pull request for the current branch, or when an existing pull request needs a better description - including when the check-description CI check has failed.

    187 GitHub stars~1.5k tokensUpdated today
    DevelopmentAuto-check passed
  • Gh Stack

    dotnet/macios

    Official

    Manage stacked branches and pull requests with the gh-stack GitHub CLI extension.

    2.9k GitHub starsUsed in 3 repos~10k tokens
    DevelopmentAuto-check passed
  • Pull Request

    OpenCoreMMO/OpenCoreMMO

    Create a pull request for OpenCoreMMO following the project's PR template.

    481 GitHub stars~814 tokensUpdated 15 days ago
    DevelopmentAuto-check passed
  • Maintain DisCatSharp

    Aiko-IT-Systems/DisCatSharp

    Guides changes to the DisCatSharp C# Discord library: tracing a payload field through parsing, serialization and caches, then validating across target frameworks.

    140 GitHub stars~1.2k tokensUpdated 2 days ago
    DevelopmentAuto-check passed

More from dotnet/maui

All 27 skills in this repo
  • Mines local Copilot CLI session logs for dotnet/maui to rank costly or failing runs, tag recurring failure modes, propose repo edits and emit guard evals.

    23k GitHub stars~3.4k tokensUpdated today
    Auto-check passed
  • Official

    Reviews the tests added in a pull request for fix coverage, quality, edge cases and test type, and recommends lighter test types where they would do.

    23k GitHub stars~2.9k tokensUpdated today
    Auto-check passed
  • Official

    Produces evidence-backed ship-readiness verdicts for .NET MAUI Servicing Releases and Previews, and drafts public-safe release handoff pages from the result.

    23k GitHub stars~15k tokensUpdated today
    Auto-check passed
  • Official

    Interprets pinned managed benchmark evidence for a dotnet/maui pull request and writes a narrative for the performance review workflow, without running or publishing anything.

    23k GitHub stars~2.4k tokensUpdated today
    Auto-check passed
  • PR Finalize

    dotnet/maui

    Official

    Checks that a pull request's title and description match its implementation and reviews the code for best practices before merge, without posting anything.

    23k GitHub stars~3.1k tokensUpdated today
    Auto-check passed
  • Official

    Adds MAUI-specific guardrails on top of the maestro-cli skill and Maestro MCP tools for darc, BAR, and channel or feed lookups in dotnet/maui.

    23k GitHub stars~10k tokensUpdated today
    Auto-check passed

Works with

Questions about Agentic Labeler

What does Agentic Labeler do?

Labels issues and pull requests in the dotnet/maui repository with area- and platform/ labels ONLY, based on technical content and platform-file conventions. Agentic Labeler is an agent skill from dotnet/maui, published by the product's own GitHub organization. Labels issues and pull requests in the dotnet/maui repository with area- and platform/ labels ONLY, based on technical content and platform-file conventions.

When should I use Agentic Labeler?

Agentic Labeler fits situations like: tasks that involve Pull requests.

How do I install Agentic Labeler in Claude Code?

Run `npx skills add dotnet/maui --skill agentic-labeler -a claude-code`. Or copy the skill folder (.github/skills/agentic-labeler in dotnet/maui) into .claude/skills/agentic-labeler in your project. Claude Code loads it when a task matches its description.

How do I install Agentic Labeler in Codex?

Run `npx skills add dotnet/maui --skill agentic-labeler -a codex`. Or copy the skill folder (.github/skills/agentic-labeler in dotnet/maui) into .agents/skills/agentic-labeler in your project. Codex loads it when a task matches its description.

Can I use Agentic Labeler in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add dotnet/maui --skill agentic-labeler -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/agentic-labeler, .gemini/skills/agentic-labeler, .github/skills/agentic-labeler and .opencode/skills/agentic-labeler in your project.

What does Agentic Labeler need to run?

SKILL.md names no scripts, command-line tools or credentials: Agentic Labeler is instructions for the agent only.

Does Agentic Labeler access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Agentic Labeler safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Agentic Labeler use?

Agentic Labeler is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Agentic Labeler use?

About 3.5k tokens (SKILL.md is roughly 14k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Agentic Labeler?

Skills that share tags, products or a category with Agentic Labeler: .NET MAUI Code Review (dotnet/efcore, 15k stars), Code Review (jonathanpeppers/dotnes, 780 stars), Open Pull Request (Gremlinq/ExRam.Gremlinq, 187 stars) and Gh Stack (dotnet/macios, 2.9k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Agentic Labeler?

dotnet (a GitHub organization, an official publisher) maintains it in dotnet/maui, which has 23,321 GitHub stars. The repository holds 27 skills in this directory. The repository was last updated on October 8, 2026.

Source: dotnet/maui on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.