{"id":"e36cf557-9362-495f-b157-d2a9a4c4aff8","shortId":"pajFty","kind":"skill","title":"microsoft-skill-creator","tagline":"Create agent skills for Microsoft technologies using Learn MCP tools. Use when users want to create a skill that teaches agents about any Microsoft technology, library, framework, or service (Azure, .NET, M365, VS Code, Bicep, etc.). Investigates topics deeply, then generates a h","description":"# Microsoft Skill Creator\n\nCreate hybrid skills for Microsoft technologies that store essential knowledge locally while enabling dynamic Learn MCP lookups for deeper details.\n\n## About Skills\n\nSkills are modular packages that extend agent capabilities with specialized knowledge and workflows. A skill transforms a general-purpose agent into a specialized one for a specific domain.\n\n### Skill Structure\n\n```\nskill-name/\n├── SKILL.md (required)     # Frontmatter (name, description) + instructions\n├── references/             # Documentation loaded into context as needed\n├── sample_codes/           # Working code examples\n└── assets/                 # Files used in output (templates, etc.)\n```\n\n### Key Principles\n\n- **Frontmatter is critical**: `name` and `description` determine when the skill triggers—be clear and comprehensive\n- **Concise is key**: Only include what agents don't already know; context window is shared\n- **No duplication**: Information lives in SKILL.md OR reference files, not both\n\n## Learn MCP Tools\n\n| Tool | Purpose | When to Use |\n|------|---------|-------------|\n| `microsoft_docs_search` | Search official docs | First pass discovery, finding topics |\n| `microsoft_docs_fetch` | Get full page content | Deep dive into important pages |\n| `microsoft_code_sample_search` | Find code examples | Get implementation patterns |\n\n### CLI Alternative\n\nIf the Learn MCP server is not available, use the `mslearn` CLI from a terminal or shell (for example, Bash, PowerShell, or cmd) instead:\n\n```bash\n# Run directly (no install needed)\nnpx @microsoft/learn-cli search \"semantic kernel overview\"\n\n# Or install globally, then run\nnpm install -g @microsoft/learn-cli\nmslearn search \"semantic kernel overview\"\n```\n\n| MCP Tool | CLI Command |\n|----------|-------------|\n| `microsoft_docs_search(query: \"...\")` | `mslearn search \"...\"` |\n| `microsoft_code_sample_search(query: \"...\", language: \"...\")` | `mslearn code-search \"...\" --language ...` |\n| `microsoft_docs_fetch(url: \"...\")` | `mslearn fetch \"...\"` |\n\nGenerated skills should include this same CLI fallback table so agents can use either path.\n\n## Creation Process\n\n### Step 1: Investigate the Topic\n\nBuild deep understanding using Learn MCP tools in three phases:\n\n**Phase 1 - Scope Discovery:**\n```\nmicrosoft_docs_search(query=\"{technology} overview what is\")\nmicrosoft_docs_search(query=\"{technology} concepts architecture\")\nmicrosoft_docs_search(query=\"{technology} getting started tutorial\")\n```\n\n**Phase 2 - Core Content:**\n```\nmicrosoft_docs_fetch(url=\"...\")  # Fetch pages from Phase 1\nmicrosoft_code_sample_search(query=\"{technology}\", language=\"{lang}\")\n```\n\n**Phase 3 - Depth:**\n```\nmicrosoft_docs_search(query=\"{technology} best practices\")\nmicrosoft_docs_search(query=\"{technology} troubleshooting errors\")\n```\n\n#### Investigation Checklist\n\nAfter investigating, verify:\n- [ ] Can explain what the technology does in one paragraph\n- [ ] Identified 3-5 key concepts\n- [ ] Have working code for basic usage\n- [ ] Know the most common API patterns\n- [ ] Have search queries for deeper topics\n\n### Step 2: Clarify with User\n\nPresent findings and ask:\n1. \"I found these key areas: [list]. Which are most important?\"\n2. \"What tasks will agents primarily perform with this skill?\"\n3. \"Which programming language should code samples prioritize?\"\n\n### Step 3: Generate the Skill\n\nUse the appropriate template from [skill-templates.md](references/skill-templates.md):\n\n| Technology Type | Template |\n|-----------------|----------|\n| Client library, NuGet/npm package | SDK/Library |\n| Azure resource | Azure Service |\n| App development framework | Framework/Platform |\n| REST API, protocol | API/Protocol |\n\n#### Generated Skill Structure\n\n```\n{skill-name}/\n├── SKILL.md                    # Core knowledge + Learn MCP guidance\n├── references/                 # Detailed local documentation (if needed)\n└── sample_codes/               # Working code examples\n    ├── getting-started/\n    └── common-patterns/\n```\n\n### Step 4: Balance Local vs Dynamic Content\n\n**Store locally when:**\n- Foundational (needed for any task)\n- Frequently accessed\n- Stable (won't change)\n- Hard to find via search\n\n**Keep dynamic when:**\n- Exhaustive reference (too large)\n- Version-specific\n- Situational (specific tasks only)\n- Well-indexed (easy to search)\n\n#### Content Guidelines\n\n| Content Type | Local | Dynamic |\n|--------------|-------|---------|\n| Core concepts (3-5) | ✅ Full | |\n| Hello world code | ✅ Full | |\n| Common patterns (3-5) | ✅ Full | |\n| Top API methods | Signature + example | Full docs via fetch |\n| Best practices | Top 5 bullets | Search for more |\n| Troubleshooting | | Search queries |\n| Full API reference | | Doc links |\n\n### Step 5: Validate\n\n1. Review: Is local content sufficient for common tasks?\n2. Test: Do suggested search queries return useful results?\n3. Verify: Do code samples run without errors?\n\n## Common Investigation Patterns\n\n### For SDKs/Libraries\n```\n\"{name} overview\" → purpose, architecture\n\"{name} getting started quickstart\" → setup steps\n\"{name} API reference\" → core classes/methods\n\"{name} samples examples\" → code patterns\n\"{name} best practices performance\" → optimization\n```\n\n### For Azure Services\n```\n\"{service} overview features\" → capabilities\n\"{service} quickstart {language}\" → setup code\n\"{service} REST API reference\" → endpoints\n\"{service} SDK {language}\" → client library\n\"{service} pricing limits quotas\" → constraints\n```\n\n### For Frameworks/Platforms\n```\n\"{framework} architecture concepts\" → mental model\n\"{framework} project structure\" → conventions\n\"{framework} tutorial walkthrough\" → end-to-end flow\n\"{framework} configuration options\" → customization\n```\n\n## Example: Creating a \"Semantic Kernel\" Skill\n\n### Investigation\n\n```\nmicrosoft_docs_search(query=\"semantic kernel overview\")\nmicrosoft_docs_search(query=\"semantic kernel plugins functions\")\nmicrosoft_code_sample_search(query=\"semantic kernel\", language=\"csharp\")\nmicrosoft_docs_fetch(url=\"https://learn.microsoft.com/semantic-kernel/overview/\")\n```\n\n### Generated Skill\n\n```\nsemantic-kernel/\n├── SKILL.md\n└── sample_codes/\n    ├── getting-started/\n    │   └── hello-kernel.cs\n    └── common-patterns/\n        ├── chat-completion.cs\n        └── function-calling.cs\n```\n\n### Generated SKILL.md\n\n```markdown\n---\nname: semantic-kernel\ndescription: Build AI agents with Microsoft Semantic Kernel. Use for LLM-powered apps with plugins, planners, and memory in .NET or Python.\n---\n\n# Semantic Kernel\n\nOrchestration SDK for integrating LLMs into applications with plugins, planners, and memory.\n\n## Key Concepts\n\n- **Kernel**: Central orchestrator managing AI services and plugins\n- **Plugins**: Collections of functions the AI can call\n- **Planner**: Sequences plugin functions to achieve goals\n- **Memory**: Vector store integration for RAG patterns\n\n## Quick Start\n\nSee [getting-started/hello-kernel.cs](sample_codes/getting-started/hello-kernel.cs)\n\n## Learn More\n\n| Topic | How to Find |\n|-------|-------------|\n| Plugin development | `microsoft_docs_search(query=\"semantic kernel plugins custom functions\")` |\n| Planners | `microsoft_docs_search(query=\"semantic kernel planner\")` |\n| Memory | `microsoft_docs_fetch(url=\"https://learn.microsoft.com/en-us/semantic-kernel/frameworks/agent/agent-memory\")` |\n\n## CLI Alternative\n\nIf the Learn MCP server is not available, use the `mslearn` CLI instead:\n\n| MCP Tool | CLI Command |\n|----------|-------------|\n| `microsoft_docs_search(query: \"...\")` | `mslearn search \"...\"` |\n| `microsoft_code_sample_search(query: \"...\", language: \"...\")` | `mslearn code-search \"...\" --language ...` |\n| `microsoft_docs_fetch(url: \"...\")` | `mslearn fetch \"...\"` |\n\nRun directly with `npx @microsoft/learn-cli <command>` or install globally with `npm install -g @microsoft/learn-cli`.\n```","tags":["microsoft","skill","creator","awesome","copilot","github","agent-skills","agents","custom-agents","github-copilot","hacktoberfest","prompt-engineering"],"capabilities":["skill","source-github","skill-microsoft-skill-creator","topic-agent-skills","topic-agents","topic-awesome","topic-custom-agents","topic-github-copilot","topic-hacktoberfest","topic-prompt-engineering"],"categories":["awesome-copilot"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/github/awesome-copilot/microsoft-skill-creator","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add github/awesome-copilot","source_repo":"https://github.com/github/awesome-copilot","install_from":"skills.sh"}},"qualityScore":"0.700","qualityRationale":"deterministic score 0.70 from registry signals: · indexed on github topic:agent-skills · 33270 github stars · SKILL.md body (7,816 chars)","verified":false,"liveness":"unknown","lastLivenessCheck":null,"agentReviews":{"count":0,"score_avg":null,"cost_usd_avg":null,"success_rate":null,"latency_p50_ms":null,"narrative_summary":null,"summary_updated_at":null},"enrichmentModel":"deterministic:skill-github:v1","enrichmentVersion":1,"enrichedAt":"2026-05-18T18:52:17.222Z","embedding":null,"createdAt":"2026-04-18T20:25:48.099Z","updatedAt":"2026-05-18T18:52:17.222Z","lastSeenAt":"2026-05-18T18:52:17.222Z","tsv":"'-5':408,583,592 '/en-us/semantic-kernel/frameworks/agent/agent-memory':900 '/hello-kernel.cs':865 '/semantic-kernel/overview/':765 '1':313,328,366,438,622 '2':355,430,449,631 '3':376,407,459,468,582,591,640 '4':529 '5':606,620 'access':544 'achiev':850 'agent':6,25,79,93,155,305,453,793 'ai':792,833,842 'alreadi':158 'altern':217,902 'api':421,496,595,615,664,692 'api/protocol':498 'app':491,803 'applic':821 'appropri':474 'architectur':345,656,708 'area':443 'ask':437 'asset':125 'avail':225,910 'azur':34,487,489,679 'balanc':530 'bash':237,242 'basic':415 'best':383,603,674 'bicep':39 'build':317,791 'bullet':607 'call':844 'capabl':80,684 'central':830 'chang':548 'chat-completion.cs':781 'checklist':393 'clarifi':431 'classes/methods':667 'clear':146 'cli':216,229,270,301,901,914,918 'client':482,698 'cmd':240 'code':38,121,123,207,211,279,286,368,413,464,518,520,587,643,671,689,751,773,927,934 'code-search':285,933 'codes/getting-started/hello-kernel.cs':867 'collect':838 'command':271,919 'common':420,526,589,629,648,779 'common-pattern':525,778 'comprehens':148 'concept':344,410,581,709,828 'concis':149 'configur':725 'constraint':704 'content':200,357,534,574,576,626 'context':117,160 'convent':715 'core':356,506,580,666 'creat':5,20,51,729 'creation':310 'creator':4,50 'critic':136 'csharp':758 'custom':727,883 'deep':201,318 'deeper':69,427 'deepli':43 'depth':377 'descript':111,139,790 'detail':70,512 'determin':140 'develop':492,875 'direct':244,944 'discoveri':191,330 'dive':202 'doc':184,188,195,273,290,332,340,347,359,379,386,600,617,736,743,760,877,887,895,921,938 'document':114,514 'domain':101 'duplic':165 'dynam':64,533,555,579 'easi':571 'either':308 'enabl':63 'end':720,722 'end-to-end':719 'endpoint':694 'error':391,647 'essenti':59 'etc':40,131 'exampl':124,212,236,521,598,670,728 'exhaust':557 'explain':398 'extend':78 'fallback':302 'featur':683 'fetch':196,291,294,360,362,602,761,896,939,942 'file':126,172 'find':192,210,435,551,873 'first':189 'flow':723 'found':440 'foundat':538 'framework':31,493,707,712,716,724 'framework/platform':494 'frameworks/platforms':706 'frequent':543 'frontmatt':109,134 'full':198,584,588,593,599,614 'function':749,840,848,884 'function-calling.cs':782 'g':261,954 'general':91 'general-purpos':90 'generat':45,295,469,499,766,783 'get':197,213,351,523,658,775,863 'getting-start':522,774,862 'global':256,950 'goal':851 'guidanc':510 'guidelin':575 'h':47 'hard':549 'hello':585 'hello-kernel.cs':777 'hybrid':52 'identifi':406 'implement':214 'import':204,448 'includ':153,298 'index':570 'inform':166 'instal':246,255,260,949,953 'instead':241,915 'instruct':112 'integr':818,855 'investig':41,314,392,395,649,734 'keep':554 'kernel':252,266,732,740,747,756,770,789,797,814,829,881,891 'key':132,151,409,442,827 'know':159,417 'knowledg':60,83,507 'lang':374 'languag':283,288,373,462,687,697,757,931,936 'larg':560 'learn':12,65,175,220,321,508,868,905 'learn.microsoft.com':764,899 'learn.microsoft.com/en-us/semantic-kernel/frameworks/agent/agent-memory':898 'learn.microsoft.com/semantic-kernel/overview/':763 'librari':30,483,699 'limit':702 'link':618 'list':444 'live':167 'llm':801 'llm-power':800 'llms':819 'load':115 'local':61,513,531,536,578,625 'lookup':67 'm365':36 'manag':832 'markdown':785 'mcp':13,66,176,221,268,322,509,906,916 'memori':808,826,852,893 'mental':710 'method':596 'microsoft':2,9,28,48,55,183,194,206,272,278,289,331,339,346,358,367,378,385,735,742,750,759,795,876,886,894,920,926,937 'microsoft-skill-cr':1 'microsoft/learn-cli':249,262,947,955 'model':711 'modular':75 'mslearn':228,263,276,284,293,913,924,932,941 'name':106,110,137,504,653,657,663,668,673,786 'need':119,247,516,539 'net':35,810 'npm':259,952 'npx':248,946 'nuget/npm':484 'offici':187 'one':97,404 'optim':677 'option':726 'orchestr':815,831 'output':129 'overview':253,267,336,654,682,741 'packag':76,485 'page':199,205,363 'paragraph':405 'pass':190 'path':309 'pattern':215,422,527,590,650,672,780,858 'perform':455,676 'phase':326,327,354,365,375 'planner':806,824,845,885,892 'plugin':748,805,823,836,837,847,874,882 'power':802 'powershel':238 'practic':384,604,675 'present':434 'price':701 'primarili':454 'principl':133 'priorit':466 'process':311 'program':461 'project':713 'protocol':497 'purpos':92,179,655 'python':812 'queri':275,282,334,342,349,371,381,388,425,613,636,738,745,754,879,889,923,930 'quick':859 'quickstart':660,686 'quota':703 'rag':857 'refer':113,171,511,558,616,665,693 'references/skill-templates.md':478 'requir':108 'resourc':488 'rest':495,691 'result':639 'return':637 'review':623 'run':243,258,645,943 'sampl':120,208,280,369,465,517,644,669,752,772,866,928 'scope':329 'sdk':696,816 'sdk/library':486 'sdks/libraries':652 'search':185,186,209,250,264,274,277,281,287,333,341,348,370,380,387,424,553,573,608,612,635,737,744,753,878,888,922,925,929,935 'see':861 'semant':251,265,731,739,746,755,769,788,796,813,880,890 'semantic-kernel':768,787 'sequenc':846 'server':222,907 'servic':33,490,680,681,685,690,695,700,834 'setup':661,688 'share':163 'shell':234 'signatur':597 'situat':564 'skill':3,7,22,49,53,72,73,87,102,105,143,296,458,471,500,503,733,767 'skill-microsoft-skill-creator' 'skill-nam':104,502 'skill-templates.md':477 'skill.md':107,169,505,771,784 'source-github' 'special':82,96 'specif':100,563,565 'stabl':545 'start':352,524,659,776,860,864 'step':312,429,467,528,619,662 'store':58,535,854 'structur':103,501,714 'suffici':627 'suggest':634 'tabl':303 'task':451,542,566,630 'teach':24 'technolog':10,29,56,335,343,350,372,382,389,401,479 'templat':130,475,481 'termin':232 'test':632 'three':325 'tool':14,177,178,269,323,917 'top':594,605 'topic':42,193,316,428,870 'topic-agent-skills' 'topic-agents' 'topic-awesome' 'topic-custom-agents' 'topic-github-copilot' 'topic-hacktoberfest' 'topic-prompt-engineering' 'transform':88 'trigger':144 'troubleshoot':390,611 'tutori':353,717 'type':480,577 'understand':319 'url':292,361,762,897,940 'usag':416 'use':11,15,127,182,226,307,320,472,638,798,911 'user':17,433 'valid':621 'vector':853 'verifi':396,641 'version':562 'version-specif':561 'via':552,601 'vs':37,532 'walkthrough':718 'want':18 'well':569 'well-index':568 'window':161 'without':646 'won':546 'work':122,412,519 'workflow':85 'world':586","prices":[{"id":"57456b68-83c5-448e-94e7-c549e0b9c46d","listingId":"e36cf557-9362-495f-b157-d2a9a4c4aff8","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"github","category":"awesome-copilot","install_from":"skills.sh"},"createdAt":"2026-04-18T20:25:48.099Z"}],"sources":[{"listingId":"e36cf557-9362-495f-b157-d2a9a4c4aff8","source":"github","sourceId":"github/awesome-copilot/microsoft-skill-creator","sourceUrl":"https://github.com/github/awesome-copilot/tree/main/skills/microsoft-skill-creator","isPrimary":false,"firstSeenAt":"2026-04-18T21:50:14.207Z","lastSeenAt":"2026-05-18T18:52:17.222Z"},{"listingId":"e36cf557-9362-495f-b157-d2a9a4c4aff8","source":"skills_sh","sourceId":"github/awesome-copilot/microsoft-skill-creator","sourceUrl":"https://skills.sh/github/awesome-copilot/microsoft-skill-creator","isPrimary":true,"firstSeenAt":"2026-04-18T20:25:48.099Z","lastSeenAt":"2026-05-07T22:40:18.108Z"}],"details":{"listingId":"e36cf557-9362-495f-b157-d2a9a4c4aff8","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"github","slug":"microsoft-skill-creator","github":{"repo":"github/awesome-copilot","stars":33270,"topics":["agent-skills","agents","ai","awesome","custom-agents","github-copilot","hacktoberfest","prompt-engineering"],"license":"mit","html_url":"https://github.com/github/awesome-copilot","pushed_at":"2026-05-18T01:26:59Z","description":"Community-contributed instructions, agents, skills, and configurations to help you make the most of GitHub Copilot.","skill_md_sha":"139faf235cf92c70317d71099e9ad7913496e6de","skill_md_path":"skills/microsoft-skill-creator/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/github/awesome-copilot/tree/main/skills/microsoft-skill-creator"},"layout":"multi","source":"github","category":"awesome-copilot","frontmatter":{"name":"microsoft-skill-creator","description":"Create agent skills for Microsoft technologies using Learn MCP tools. Use when users want to create a skill that teaches agents about any Microsoft technology, library, framework, or service (Azure, .NET, M365, VS Code, Bicep, etc.). Investigates topics deeply, then generates a hybrid skill storing essential knowledge locally while enabling dynamic deeper investigation.","compatibility":"Works best with Microsoft Learn MCP Server (https://learn.microsoft.com/api/mcp). Can also use the mslearn CLI as a fallback."},"skills_sh_url":"https://skills.sh/github/awesome-copilot/microsoft-skill-creator"},"updatedAt":"2026-05-18T18:52:17.222Z"}}