[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M69AT2N5V66CC234GHJMH0DJ":3,"skill-same:01M69AT2N5V66CC234GHJMH0DJ":106},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":38,"linkOnly":39,"markdownContent":40,"markdownBody":41,"frontMatter":42,"refName":47,"commitSha":48,"skillDirectory":49,"forksCount":50,"repositoryHtmlUrl":51,"repositoryHomepageUrl":52,"ownerHtmlUrl":53,"ownerBlogUrl":54,"ownerDescription":55,"updatedAt":56,"install":57,"resources":70},"01M69AT2N5V66CC234GHJMH0DJ","azure-search-documents-py",null,"Azure AI Search SDK for Python. Use for vector search, hybrid search, semantic ranking, indexing, and skillsets.\nTriggers: \"azure-search-documents\", \"SearchClient\", \"SearchIndexClient\", \"vector search\", \"hybrid search\", \"semantic search\".","microsoft\u002Fskills","microsoft","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F6154722?v=4","01M698VYDXZ6JW5WH3CRV650R8","Microsoft","01M69AJYYW4E1RMSYCQDJ4FJSH",3204,"MIT","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fskills\u002Fblob\u002Fmain\u002F.github\u002Fplugins\u002Fazure-sdk-python\u002Fskills\u002Fazure-search-documents-py\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,9,37],"documents","azure","azure search documents py","search","search documents","Vector Search","hybrid","for","Hybrid Search","ranking","semantic","triggers","indexing","use","python","vector","semantic search","azure ai","sdk","2026-11-10 10:12:30",false,"---\nname: azure-search-documents-py\ndescription: |\n  Azure AI Search SDK for Python. Use for vector search, hybrid search, semantic ranking, indexing, and skillsets.\n  Triggers: \"azure-search-documents\", \"SearchClient\", \"SearchIndexClient\", \"vector search\", \"hybrid search\", \"semantic search\".\nlicense: MIT\nmetadata:\n  author: Microsoft\n  version: \"1.0.0\"\n  package: azure-search-documents\n---\n\n# Azure AI Search SDK for Python\n\nFull-text, vector, and hybrid search with AI enrichment capabilities.\n\n## Installation\n\n```bash\npip install azure-search-documents\n```\n\n## Environment Variables\n\n```bash\nAZURE_SEARCH_ENDPOINT=https:\u002F\u002F\u003Cservice-name>.search.windows.net  # Required for all auth methods\nAZURE_SEARCH_INDEX_NAME=\u003Cyour-index-name>  # Required for all auth methods\nAZURE_TOKEN_CREDENTIALS=prod # Required only if DefaultAzureCredential is used in production\nAZURE_SEARCH_API_KEY=\u003Cyour-api-key>  # Only required for the legacy API-key auth path below\n```\n\n## Authentication & Lifecycle\n\n> **? Two rules apply to every code sample below:**\n>\n> 1. **Prefer `DefaultAzureCredential`.** It works locally (Azure CLI \u002F VS Code \u002F Developer CLI) and in Azure (managed identity, workload identity) with no code change. Avoid connection strings, account\u002FAPI keys — they bypass Entra audit and rotation.\n>    - Local dev: `DefaultAzureCredential` works as-is.\n>    - Production: set `AZURE_TOKEN_CREDENTIALS=prod` (or `AZURE_TOKEN_CREDENTIALS=\u003Cspecific_credential>`) to constrain the credential chain to production-safe credentials.\n> 2. **Wrap every client in a context manager** so HTTP transports, sockets, and token caches are released deterministically:\n>    - Sync: `with \u003CClient>(...) as client:`\n>    - Async: `async with \u003CClient>(...) as client:` **and** `async with DefaultAzureCredential() as credential:` (from `azure.identity.aio`)\n>\n> Snippets may abbreviate this setup, but production code should always follow both rules.\n\n```python\nimport os\nfrom azure.identity import DefaultAzureCredential, ManagedIdentityCredential\nfrom azure.search.documents import SearchClient\n\n# Local dev: DefaultAzureCredential. Production: set AZURE_TOKEN_CREDENTIALS=prod or AZURE_TOKEN_CREDENTIALS=\u003Cspecific_credential>\ncredential = DefaultAzureCredential(require_envvar=True)\n# Or use a specific credential directly in production:\n# See https:\u002F\u002Flearn.microsoft.com\u002Fpython\u002Fapi\u002Foverview\u002Fazure\u002Fidentity-readme?view=azure-python#credential-classes\n# credential = ManagedIdentityCredential()\n\nwith SearchClient(\n    endpoint=os.environ[\"AZURE_SEARCH_ENDPOINT\"],\n    index_name=os.environ[\"AZURE_SEARCH_INDEX_NAME\"],\n    credential=credential,\n) as client:\n    results = list(client.search(search_text=\"*\", top=5))\n```\n\n### Legacy: API Key (existing keyed deployments)\n\nNew code should use `DefaultAzureCredential` above. Use `AzureKeyCredential` only if you have an existing keyed deployment that hasn't been migrated to Entra ID yet — for example, regulated environments still completing their Entra rollout. The same `AzureKeyCredential` works with `SearchIndexClient` and `SearchIndexerClient` for admin operations.\n\n```python\nimport os\nfrom azure.core.credentials import AzureKeyCredential\nfrom azure.search.documents import SearchClient\n\nwith SearchClient(\n    endpoint=os.environ[\"AZURE_SEARCH_ENDPOINT\"],\n    index_name=os.environ[\"AZURE_SEARCH_INDEX_NAME\"],\n    credential=AzureKeyCredential(os.environ[\"AZURE_SEARCH_API_KEY\"]),\n) as client:\n    results = list(client.search(search_text=\"*\", top=5))\n```\n\n## Client Types\n\n| Client | Purpose |\n|--------|---------|\n| `SearchClient` | Search and document operations |\n| `SearchIndexClient` | Index management, synonym maps |\n| `SearchIndexerClient` | Indexers, data sources, skillsets |\n\n## Create Index with Vector Field\n\n```python\nfrom azure.search.documents.indexes import SearchIndexClient\nfrom azure.search.documents.indexes.models import (\n    SearchIndex,\n    SearchField,\n    SearchFieldDataType,\n    VectorSearch,\n    HnswAlgorithmConfiguration,\n    VectorSearchProfile,\n    SearchableField,\n    SimpleField\n)\n\nfields = [\n    SimpleField(name=\"id\", type=SearchFieldDataType.String, key=True),\n    SearchableField(name=\"title\", type=SearchFieldDataType.String),\n    SearchableField(name=\"content\", type=SearchFieldDataType.String),\n    SearchField(\n        name=\"content_vector\",\n        type=SearchFieldDataType.Collection(SearchFieldDataType.Single),\n        searchable=True,\n        vector_search_dimensions=1536,\n        vector_search_profile_name=\"my-vector-profile\"\n    )\n]\n\nvector_search = VectorSearch(\n    algorithms=[\n        HnswAlgorithmConfiguration(name=\"my-hnsw\")\n    ],\n    profiles=[\n        VectorSearchProfile(\n            name=\"my-vector-profile\",\n            algorithm_configuration_name=\"my-hnsw\"\n        )\n    ]\n)\n\nindex = SearchIndex(\n    name=\"my-index\",\n    fields=fields,\n    vector_search=vector_search\n)\n\nwith SearchIndexClient(endpoint, DefaultAzureCredential()) as index_client:\n    index_client.create_or_update_index(index)\n```\n\n## Upload Documents\n\n```python\nfrom azure.search.documents import SearchClient\n\ndocuments = [\n    {\n        \"id\": \"1\",\n        \"title\": \"Azure AI Search\",\n        \"content\": \"Full-text and vector search service\",\n        \"content_vector\": [0.1, 0.2, ...]  # 1536 dimensions\n    }\n]\n\nwith SearchClient(endpoint, \"my-index\", DefaultAzureCredential()) as client:\n    result = client.upload_documents(documents)\n    print(f\"Uploaded {len(result)} documents\")\n```\n\n## Keyword Search\n\n```python\nresults = client.search(\n    search_text=\"azure search\",\n    select=[\"id\", \"title\", \"content\"],\n    top=10\n)\n\nfor result in results:\n    print(f\"{result['title']}: {result['@search.score']}\")\n```\n\n## Vector Search\n\n```python\nfrom azure.search.documents.models import VectorizedQuery\n\n# Your query embedding (1536 dimensions)\nquery_vector = get_embedding(\"semantic search capabilities\")\n\nvector_query = VectorizedQuery(\n    vector=query_vector,\n    k_nearest_neighbors=10,\n    fields=\"content_vector\"\n)\n\nresults = client.search(\n    vector_queries=[vector_query],\n    select=[\"id\", \"title\", \"content\"]\n)\n\nfor result in results:\n    print(f\"{result['title']}: {result['@search.score']}\")\n```\n\n## Hybrid Search (Vector + Keyword)\n\n```python\nfrom azure.search.documents.models import VectorizedQuery\n\nvector_query = VectorizedQuery(\n    vector=query_vector,\n    k_nearest_neighbors=10,\n    fields=\"content_vector\"\n)\n\nresults = client.search(\n    search_text=\"azure search\",\n    vector_queries=[vector_query],\n    select=[\"id\", \"title\", \"content\"],\n    top=10\n)\n```\n\n## Semantic Ranking\n\n```python\nfrom azure.search.documents.models import QueryType\n\nresults = client.search(\n    search_text=\"what is azure search\",\n    query_type=QueryType.SEMANTIC,\n    semantic_configuration_name=\"my-semantic-config\",\n    select=[\"id\", \"title\", \"content\"],\n    top=10\n)\n\nfor result in results:\n    print(f\"{result['title']}\")\n    if result.get(\"@search.captions\"):\n        print(f\"  Caption: {result['@search.captions'][0].text}\")\n```\n\n## Filters\n\n```python\nresults = client.search(\n    search_text=\"*\",\n    filter=\"category eq 'Technology' and rating gt 4\",\n    order_by=[\"rating desc\"],\n    select=[\"id\", \"title\", \"category\", \"rating\"]\n)\n```\n\n## Facets\n\n```python\nresults = client.search(\n    search_text=\"*\",\n    facets=[\"category,count:10\", \"rating\"],\n    top=0  # Only get facets, no documents\n)\n\nfor facet_name, facet_values in results.get_facets().items():\n    print(f\"{facet_name}:\")\n    for facet in facet_values:\n        print(f\"  {facet['value']}: {facet['count']}\")\n```\n\n## Autocomplete & Suggest\n\n```python\n# Autocomplete\nresults = client.autocomplete(\n    search_text=\"sea\",\n    suggester_name=\"my-suggester\",\n    mode=\"twoTerms\"\n)\n\n# Suggest\nresults = client.suggest(\n    search_text=\"sea\",\n    suggester_name=\"my-suggester\",\n    select=[\"title\"]\n)\n```\n\n## Indexer with Skillset\n\n```python\nfrom azure.search.documents.indexes import SearchIndexerClient\nfrom azure.search.documents.indexes.models import (\n    SearchIndexer,\n    SearchIndexerDataSourceConnection,\n    SearchIndexerSkillset,\n    EntityRecognitionSkill,\n    InputFieldMappingEntry,\n    OutputFieldMappingEntry\n)\n\nwith SearchIndexerClient(endpoint, DefaultAzureCredential()) as indexer_client:\n    # Use managed identity (search service must have RBAC role on the storage account). Avoid storage connection strings with embedded keys.\n    data_source = SearchIndexerDataSourceConnection(\n        name=\"my-datasource\",\n        type=\"azureblob\",\n        connection_string=\"ResourceId=\u002Fsubscriptions\u002F\u003Csub>\u002FresourceGroups\u002F\u003Crg>\u002Fproviders\u002FMicrosoft.Storage\u002FstorageAccounts\u002F\u003Cacct>\",\n        container={\"name\": \"documents\"}\n    )\n    indexer_client.create_or_update_data_source_connection(data_source)\n\n    # Create skillset\n    skillset = SearchIndexerSkillset(\n        name=\"my-skillset\",\n        skills=[\n            EntityRecognitionSkill(\n                inputs=[InputFieldMappingEntry(name=\"text\", source=\"\u002Fdocument\u002Fcontent\")],\n                outputs=[OutputFieldMappingEntry(name=\"organizations\", target_name=\"organizations\")]\n            )\n        ]\n    )\n    indexer_client.create_or_update_skillset(skillset)\n\n    # Create indexer\n    indexer = SearchIndexer(\n        name=\"my-indexer\",\n        data_source_name=\"my-datasource\",\n        target_index_name=\"my-index\",\n        skillset_name=\"my-skillset\"\n    )\n    indexer_client.create_or_update_indexer(indexer)\n```\n\n## Best Practices\n\n1. **Pick sync OR async and stay consistent.** Do not mix `azure.xxx` sync clients with `azure.xxx.aio` async clients in the same call path. Choose one mode per module.\n2. **Always use context managers for clients and async credentials.** Wrap every client in `with Client(...) as client:` (sync) or `async with Client(...) as client:` (async). For async `DefaultAzureCredential` from `azure.identity.aio`, also use `async with credential:` so tokens and transports are cleaned up.\n3. **Use hybrid search** for best relevance combining vector and keyword\n4. **Enable semantic ranking** for natural language queries\n5. **Index in batches** of 100-1000 documents for efficiency\n6. **Use filters** to narrow results before ranking\n7. **Configure vector dimensions** to match your embedding model\n8. **Use HNSW algorithm** for large-scale vector search\n9. **Create suggesters** at index creation time (cannot add later)\n\n## Reference Files\n\n| File | Contents |\n|------|----------|\n| [references\u002Fvector-search.md](references\u002Fvector-search.md) | HNSW configuration, integrated vectorization, multi-vector queries |\n| [references\u002Fsemantic-ranking.md](references\u002Fsemantic-ranking.md) | Semantic configuration, captions, answers, hybrid patterns |\n| [scripts\u002Fsetup_vector_index.py](scripts\u002Fsetup_vector_index.py) | CLI script to create vector-enabled search index |\n\n\n---\n\n## Additional Azure AI Search Patterns\n\n# Azure AI Search Python SDK\n\nWrite clean, idiomatic Python code for Azure AI Search using `azure-search-documents`.\n\n## Installation\n\n```bash\npip install azure-search-documents azure-identity\n```\n\n## Environment Variables\n\n```bash\nAZURE_SEARCH_ENDPOINT=https:\u002F\u002F\u003Csearch-service>.search.windows.net  # Required for all auth methods\nAZURE_SEARCH_INDEX_NAME=\u003Cindex-name>  # Required for all auth methods\nAZURE_TOKEN_CREDENTIALS=prod # Required only if DefaultAzureCredential is used in production\n```\n\n## Authentication\n\n```python\nimport os\nfrom azure.identity import DefaultAzureCredential, ManagedIdentityCredential\nfrom azure.search.documents import SearchClient\n\n# Local dev: DefaultAzureCredential. Production: set AZURE_TOKEN_CREDENTIALS=prod or AZURE_TOKEN_CREDENTIALS=\u003Cspecific_credential>\ncredential = DefaultAzureCredential(require_envvar=True)\n# Or use a specific credential directly in production:\n# See https:\u002F\u002Flearn.microsoft.com\u002Fpython\u002Fapi\u002Foverview\u002Fazure\u002Fidentity-readme?view=azure-python#credential-classes\n# credential = ManagedIdentityCredential()\n\nwith SearchClient(\n    endpoint=os.environ[\"AZURE_SEARCH_ENDPOINT\"],\n    index_name=os.environ[\"AZURE_SEARCH_INDEX_NAME\"],\n    credential=credential,\n) as client:\n    results = list(client.search(search_text=\"*\", top=5))\n```\n\n## Client Selection\n\n| Client | Purpose |\n|--------|---------|\n| `SearchClient` | Query indexes, upload\u002Fupdate\u002Fdelete documents |\n| `SearchIndexClient` | Create\u002Fmanage indexes, knowledge sources, knowledge bases |\n| `SearchIndexerClient` | Manage indexers, skillsets, data sources |\n| `KnowledgeBaseRetrievalClient` | Agentic retrieval with LLM-powered Q&A |\n\n## Index Creation Pattern\n\n```python\nfrom azure.search.documents.indexes import SearchIndexClient\nfrom azure.search.documents.indexes.models import (\n    SearchIndex, SearchField, VectorSearch, VectorSearchProfile,\n    HnswAlgorithmConfiguration, AzureOpenAIVectorizer,\n    AzureOpenAIVectorizerParameters, SemanticSearch,\n    SemanticConfiguration, SemanticPrioritizedFields, SemanticField\n)\n\nindex = SearchIndex(\n    name=index_name,\n    fields=[\n        SearchField(name=\"id\", type=\"Edm.String\", key=True),\n        SearchField(name=\"content\", type=\"Edm.String\", searchable=True),\n        SearchField(name=\"embedding\", type=\"Collection(Edm.Single)\",\n                   vector_search_dimensions=3072,\n                   vector_search_profile_name=\"vector-profile\"),\n    ],\n    vector_search=VectorSearch(\n        profiles=[VectorSearchProfile(\n            name=\"vector-profile\",\n            algorithm_configuration_name=\"hnsw-algo\",\n            vectorizer_name=\"openai-vectorizer\"\n        )],\n        algorithms=[HnswAlgorithmConfiguration(name=\"hnsw-algo\")],\n        vectorizers=[AzureOpenAIVectorizer(\n            vectorizer_name=\"openai-vectorizer\",\n            parameters=AzureOpenAIVectorizerParameters(\n                resource_url=aoai_endpoint,\n                deployment_name=embedding_deployment,\n                model_name=embedding_model\n            )\n        )]\n    ),\n    semantic_search=SemanticSearch(\n        default_configuration_name=\"semantic-config\",\n        configurations=[SemanticConfiguration(\n            name=\"semantic-config\",\n            prioritized_fields=SemanticPrioritizedFields(\n                content_fields=[SemanticField(field_name=\"content\")]\n            )\n        )]\n    )\n)\n\nwith SearchIndexClient(endpoint, credential) as index_client:\n    index_client.create_or_update_index(index)\n```\n\n## Document Operations\n\n```python\nfrom azure.search.documents import SearchIndexingBufferedSender\n\n# Batch upload with automatic batching\nwith SearchIndexingBufferedSender(endpoint, index_name, credential) as sender:\n    sender.upload_documents(documents)\n\n# Direct operations via SearchClient\nwith SearchClient(endpoint, index_name, credential) as search_client:\n    search_client.upload_documents(documents)      # Add new\n    search_client.merge_documents(documents)       # Update existing\n    search_client.merge_or_upload_documents(documents)  # Upsert\n    search_client.delete_documents(documents)      # Remove\n```\n\n## Search Patterns\n\n```python\n# Basic search\nresults = search_client.search(search_text=\"query\")\n\n# Vector search\nfrom azure.search.documents.models import VectorizedQuery\n\nresults = search_client.search(\n    search_text=None,\n    vector_queries=[VectorizedQuery(\n        vector=embedding,\n        k_nearest_neighbors=5,\n        fields=\"embedding\"\n    )]\n)\n\n# Hybrid search (vector + keyword)\nresults = search_client.search(\n    search_text=\"query\",\n    vector_queries=[VectorizedQuery(vector=embedding, k_nearest_neighbors=5, fields=\"embedding\")],\n    query_type=\"semantic\",\n    semantic_configuration_name=\"semantic-config\"\n)\n\n# With filters\nresults = search_client.search(\n    search_text=\"query\",\n    filter=\"category eq 'technology'\",\n    select=[\"id\", \"title\", \"content\"],\n    top=10\n)\n```\n\n## Agentic Retrieval (Knowledge Bases)\n\nFor LLM-powered Q&A with answer synthesis, see [references\u002Fagentic-retrieval.md](references\u002Fagentic-retrieval.md).\n\nKey concepts:\n- **Knowledge Source**: Points to a search index\n- **Knowledge Base**: Wraps knowledge sources + LLM for query planning and synthesis\n- **Output modes**: `EXTRACTIVE_DATA` (raw chunks) or `ANSWER_SYNTHESIS` (LLM-generated answers)\n\n## Async Pattern\n\n```python\nfrom azure.search.documents.aio import SearchClient\n\nasync with SearchClient(endpoint, index_name, credential) as client:\n    results = await client.search(search_text=\"query\")\n    async for result in results:\n        print(result[\"title\"])\n```\n\n## Best Practices\n\n1. **Use environment variables** for endpoints, keys, and deployment names\n2. **Use `DefaultAzureCredential`** for code that runs locally (instead of API keys). Use a specific token credential for code that runs in Azure.\n3. **Use `SearchIndexingBufferedSender`** for batch uploads (handles batching\u002Fretries)\n4. **Always define semantic configuration** for agentic retrieval indexes\n5. **Use `create_or_update_index`** for idempotent index creation\n6. **Close clients** with context managers or explicit `close()`\n\n## Field Types Reference\n\n| EDM Type | Python | Notes |\n|----------|--------|-------|\n| `Edm.String` | str | Searchable text |\n| `Edm.Int32` | int | Integer |\n| `Edm.Int64` | int | Long integer |\n| `Edm.Double` | float | Floating point |\n| `Edm.Boolean` | bool | True\u002FFalse |\n| `Edm.DateTimeOffset` | datetime | ISO 8601 |\n| `Collection(Edm.Single)` | List[float] | Vector embeddings |\n| `Collection(Edm.String)` | List[str] | String arrays |\n\n## Error Handling\n\n```python\nfrom azure.core.exceptions import (\n    HttpResponseError,\n    ResourceNotFoundError,\n    ResourceExistsError\n)\n\ntry:\n    result = search_client.get_document(key=\"123\")\nexcept ResourceNotFoundError:\n    print(\"Document not found\")\nexcept HttpResponseError as e:\n    print(f\"Search error: {e.message}\")\n```\n","# Azure AI Search SDK for Python\n\nFull-text, vector, and hybrid search with AI enrichment capabilities.\n\n## Installation\n\n```bash\npip install azure-search-documents\n```\n\n## Environment Variables\n\n```bash\nAZURE_SEARCH_ENDPOINT=https:\u002F\u002F\u003Cservice-name>.search.windows.net  # Required for all auth methods\nAZURE_SEARCH_INDEX_NAME=\u003Cyour-index-name>  # Required for all auth methods\nAZURE_TOKEN_CREDENTIALS=prod # Required only if DefaultAzureCredential is used in production\nAZURE_SEARCH_API_KEY=\u003Cyour-api-key>  # Only required for the legacy API-key auth path below\n```\n\n## Authentication & Lifecycle\n\n> **? Two rules apply to every code sample below:**\n>\n> 1. **Prefer `DefaultAzureCredential`.** It works locally (Azure CLI \u002F VS Code \u002F Developer CLI) and in Azure (managed identity, workload identity) with no code change. Avoid connection strings, account\u002FAPI keys — they bypass Entra audit and rotation.\n>    - Local dev: `DefaultAzureCredential` works as-is.\n>    - Production: set `AZURE_TOKEN_CREDENTIALS=prod` (or `AZURE_TOKEN_CREDENTIALS=\u003Cspecific_credential>`) to constrain the credential chain to production-safe credentials.\n> 2. **Wrap every client in a context manager** so HTTP transports, sockets, and token caches are released deterministically:\n>    - Sync: `with \u003CClient>(...) as client:`\n>    - Async: `async with \u003CClient>(...) as client:` **and** `async with DefaultAzureCredential() as credential:` (from `azure.identity.aio`)\n>\n> Snippets may abbreviate this setup, but production code should always follow both rules.\n\n```python\nimport os\nfrom azure.identity import DefaultAzureCredential, ManagedIdentityCredential\nfrom azure.search.documents import SearchClient\n\n# Local dev: DefaultAzureCredential. Production: set AZURE_TOKEN_CREDENTIALS=prod or AZURE_TOKEN_CREDENTIALS=\u003Cspecific_credential>\ncredential = DefaultAzureCredential(require_envvar=True)\n# Or use a specific credential directly in production:\n# See https:\u002F\u002Flearn.microsoft.com\u002Fpython\u002Fapi\u002Foverview\u002Fazure\u002Fidentity-readme?view=azure-python#credential-classes\n# credential = ManagedIdentityCredential()\n\nwith SearchClient(\n    endpoint=os.environ[\"AZURE_SEARCH_ENDPOINT\"],\n    index_name=os.environ[\"AZURE_SEARCH_INDEX_NAME\"],\n    credential=credential,\n) as client:\n    results = list(client.search(search_text=\"*\", top=5))\n```\n\n### Legacy: API Key (existing keyed deployments)\n\nNew code should use `DefaultAzureCredential` above. Use `AzureKeyCredential` only if you have an existing keyed deployment that hasn't been migrated to Entra ID yet — for example, regulated environments still completing their Entra rollout. The same `AzureKeyCredential` works with `SearchIndexClient` and `SearchIndexerClient` for admin operations.\n\n```python\nimport os\nfrom azure.core.credentials import AzureKeyCredential\nfrom azure.search.documents import SearchClient\n\nwith SearchClient(\n    endpoint=os.environ[\"AZURE_SEARCH_ENDPOINT\"],\n    index_name=os.environ[\"AZURE_SEARCH_INDEX_NAME\"],\n    credential=AzureKeyCredential(os.environ[\"AZURE_SEARCH_API_KEY\"]),\n) as client:\n    results = list(client.search(search_text=\"*\", top=5))\n```\n\n## Client Types\n\n| Client | Purpose |\n|--------|---------|\n| `SearchClient` | Search and document operations |\n| `SearchIndexClient` | Index management, synonym maps |\n| `SearchIndexerClient` | Indexers, data sources, skillsets |\n\n## Create Index with Vector Field\n\n```python\nfrom azure.search.documents.indexes import SearchIndexClient\nfrom azure.search.documents.indexes.models import (\n    SearchIndex,\n    SearchField,\n    SearchFieldDataType,\n    VectorSearch,\n    HnswAlgorithmConfiguration,\n    VectorSearchProfile,\n    SearchableField,\n    SimpleField\n)\n\nfields = [\n    SimpleField(name=\"id\", type=SearchFieldDataType.String, key=True),\n    SearchableField(name=\"title\", type=SearchFieldDataType.String),\n    SearchableField(name=\"content\", type=SearchFieldDataType.String),\n    SearchField(\n        name=\"content_vector\",\n        type=SearchFieldDataType.Collection(SearchFieldDataType.Single),\n        searchable=True,\n        vector_search_dimensions=1536,\n        vector_search_profile_name=\"my-vector-profile\"\n    )\n]\n\nvector_search = VectorSearch(\n    algorithms=[\n        HnswAlgorithmConfiguration(name=\"my-hnsw\")\n    ],\n    profiles=[\n        VectorSearchProfile(\n            name=\"my-vector-profile\",\n            algorithm_configuration_name=\"my-hnsw\"\n        )\n    ]\n)\n\nindex = SearchIndex(\n    name=\"my-index\",\n    fields=fields,\n    vector_search=vector_search\n)\n\nwith SearchIndexClient(endpoint, DefaultAzureCredential()) as index_client:\n    index_client.create_or_update_index(index)\n```\n\n## Upload Documents\n\n```python\nfrom azure.search.documents import SearchClient\n\ndocuments = [\n    {\n        \"id\": \"1\",\n        \"title\": \"Azure AI Search\",\n        \"content\": \"Full-text and vector search service\",\n        \"content_vector\": [0.1, 0.2, ...]  # 1536 dimensions\n    }\n]\n\nwith SearchClient(endpoint, \"my-index\", DefaultAzureCredential()) as client:\n    result = client.upload_documents(documents)\n    print(f\"Uploaded {len(result)} documents\")\n```\n\n## Keyword Search\n\n```python\nresults = client.search(\n    search_text=\"azure search\",\n    select=[\"id\", \"title\", \"content\"],\n    top=10\n)\n\nfor result in results:\n    print(f\"{result['title']}: {result['@search.score']}\")\n```\n\n## Vector Search\n\n```python\nfrom azure.search.documents.models import VectorizedQuery\n\n# Your query embedding (1536 dimensions)\nquery_vector = get_embedding(\"semantic search capabilities\")\n\nvector_query = VectorizedQuery(\n    vector=query_vector,\n    k_nearest_neighbors=10,\n    fields=\"content_vector\"\n)\n\nresults = client.search(\n    vector_queries=[vector_query],\n    select=[\"id\", \"title\", \"content\"]\n)\n\nfor result in results:\n    print(f\"{result['title']}: {result['@search.score']}\")\n```\n\n## Hybrid Search (Vector + Keyword)\n\n```python\nfrom azure.search.documents.models import VectorizedQuery\n\nvector_query = VectorizedQuery(\n    vector=query_vector,\n    k_nearest_neighbors=10,\n    fields=\"content_vector\"\n)\n\nresults = client.search(\n    search_text=\"azure search\",\n    vector_queries=[vector_query],\n    select=[\"id\", \"title\", \"content\"],\n    top=10\n)\n```\n\n## Semantic Ranking\n\n```python\nfrom azure.search.documents.models import QueryType\n\nresults = client.search(\n    search_text=\"what is azure search\",\n    query_type=QueryType.SEMANTIC,\n    semantic_configuration_name=\"my-semantic-config\",\n    select=[\"id\", \"title\", \"content\"],\n    top=10\n)\n\nfor result in results:\n    print(f\"{result['title']}\")\n    if result.get(\"@search.captions\"):\n        print(f\"  Caption: {result['@search.captions'][0].text}\")\n```\n\n## Filters\n\n```python\nresults = client.search(\n    search_text=\"*\",\n    filter=\"category eq 'Technology' and rating gt 4\",\n    order_by=[\"rating desc\"],\n    select=[\"id\", \"title\", \"category\", \"rating\"]\n)\n```\n\n## Facets\n\n```python\nresults = client.search(\n    search_text=\"*\",\n    facets=[\"category,count:10\", \"rating\"],\n    top=0  # Only get facets, no documents\n)\n\nfor facet_name, facet_values in results.get_facets().items():\n    print(f\"{facet_name}:\")\n    for facet in facet_values:\n        print(f\"  {facet['value']}: {facet['count']}\")\n```\n\n## Autocomplete & Suggest\n\n```python\n# Autocomplete\nresults = client.autocomplete(\n    search_text=\"sea\",\n    suggester_name=\"my-suggester\",\n    mode=\"twoTerms\"\n)\n\n# Suggest\nresults = client.suggest(\n    search_text=\"sea\",\n    suggester_name=\"my-suggester\",\n    select=[\"title\"]\n)\n```\n\n## Indexer with Skillset\n\n```python\nfrom azure.search.documents.indexes import SearchIndexerClient\nfrom azure.search.documents.indexes.models import (\n    SearchIndexer,\n    SearchIndexerDataSourceConnection,\n    SearchIndexerSkillset,\n    EntityRecognitionSkill,\n    InputFieldMappingEntry,\n    OutputFieldMappingEntry\n)\n\nwith SearchIndexerClient(endpoint, DefaultAzureCredential()) as indexer_client:\n    # Use managed identity (search service must have RBAC role on the storage account). Avoid storage connection strings with embedded keys.\n    data_source = SearchIndexerDataSourceConnection(\n        name=\"my-datasource\",\n        type=\"azureblob\",\n        connection_string=\"ResourceId=\u002Fsubscriptions\u002F\u003Csub>\u002FresourceGroups\u002F\u003Crg>\u002Fproviders\u002FMicrosoft.Storage\u002FstorageAccounts\u002F\u003Cacct>\",\n        container={\"name\": \"documents\"}\n    )\n    indexer_client.create_or_update_data_source_connection(data_source)\n\n    # Create skillset\n    skillset = SearchIndexerSkillset(\n        name=\"my-skillset\",\n        skills=[\n            EntityRecognitionSkill(\n                inputs=[InputFieldMappingEntry(name=\"text\", source=\"\u002Fdocument\u002Fcontent\")],\n                outputs=[OutputFieldMappingEntry(name=\"organizations\", target_name=\"organizations\")]\n            )\n        ]\n    )\n    indexer_client.create_or_update_skillset(skillset)\n\n    # Create indexer\n    indexer = SearchIndexer(\n        name=\"my-indexer\",\n        data_source_name=\"my-datasource\",\n        target_index_name=\"my-index\",\n        skillset_name=\"my-skillset\"\n    )\n    indexer_client.create_or_update_indexer(indexer)\n```\n\n## Best Practices\n\n1. **Pick sync OR async and stay consistent.** Do not mix `azure.xxx` sync clients with `azure.xxx.aio` async clients in the same call path. Choose one mode per module.\n2. **Always use context managers for clients and async credentials.** Wrap every client in `with Client(...) as client:` (sync) or `async with Client(...) as client:` (async). For async `DefaultAzureCredential` from `azure.identity.aio`, also use `async with credential:` so tokens and transports are cleaned up.\n3. **Use hybrid search** for best relevance combining vector and keyword\n4. **Enable semantic ranking** for natural language queries\n5. **Index in batches** of 100-1000 documents for efficiency\n6. **Use filters** to narrow results before ranking\n7. **Configure vector dimensions** to match your embedding model\n8. **Use HNSW algorithm** for large-scale vector search\n9. **Create suggesters** at index creation time (cannot add later)\n\n## Reference Files\n\n| File | Contents |\n|------|----------|\n| [references\u002Fvector-search.md](references\u002Fvector-search.md) | HNSW configuration, integrated vectorization, multi-vector queries |\n| [references\u002Fsemantic-ranking.md](references\u002Fsemantic-ranking.md) | Semantic configuration, captions, answers, hybrid patterns |\n| [scripts\u002Fsetup_vector_index.py](scripts\u002Fsetup_vector_index.py) | CLI script to create vector-enabled search index |\n\n\n---\n\n## Additional Azure AI Search Patterns\n\n# Azure AI Search Python SDK\n\nWrite clean, idiomatic Python code for Azure AI Search using `azure-search-documents`.\n\n## Installation\n\n```bash\npip install azure-search-documents azure-identity\n```\n\n## Environment Variables\n\n```bash\nAZURE_SEARCH_ENDPOINT=https:\u002F\u002F\u003Csearch-service>.search.windows.net  # Required for all auth methods\nAZURE_SEARCH_INDEX_NAME=\u003Cindex-name>  # Required for all auth methods\nAZURE_TOKEN_CREDENTIALS=prod # Required only if DefaultAzureCredential is used in production\n```\n\n## Authentication\n\n```python\nimport os\nfrom azure.identity import DefaultAzureCredential, ManagedIdentityCredential\nfrom azure.search.documents import SearchClient\n\n# Local dev: DefaultAzureCredential. Production: set AZURE_TOKEN_CREDENTIALS=prod or AZURE_TOKEN_CREDENTIALS=\u003Cspecific_credential>\ncredential = DefaultAzureCredential(require_envvar=True)\n# Or use a specific credential directly in production:\n# See https:\u002F\u002Flearn.microsoft.com\u002Fpython\u002Fapi\u002Foverview\u002Fazure\u002Fidentity-readme?view=azure-python#credential-classes\n# credential = ManagedIdentityCredential()\n\nwith SearchClient(\n    endpoint=os.environ[\"AZURE_SEARCH_ENDPOINT\"],\n    index_name=os.environ[\"AZURE_SEARCH_INDEX_NAME\"],\n    credential=credential,\n) as client:\n    results = list(client.search(search_text=\"*\", top=5))\n```\n\n## Client Selection\n\n| Client | Purpose |\n|--------|---------|\n| `SearchClient` | Query indexes, upload\u002Fupdate\u002Fdelete documents |\n| `SearchIndexClient` | Create\u002Fmanage indexes, knowledge sources, knowledge bases |\n| `SearchIndexerClient` | Manage indexers, skillsets, data sources |\n| `KnowledgeBaseRetrievalClient` | Agentic retrieval with LLM-powered Q&A |\n\n## Index Creation Pattern\n\n```python\nfrom azure.search.documents.indexes import SearchIndexClient\nfrom azure.search.documents.indexes.models import (\n    SearchIndex, SearchField, VectorSearch, VectorSearchProfile,\n    HnswAlgorithmConfiguration, AzureOpenAIVectorizer,\n    AzureOpenAIVectorizerParameters, SemanticSearch,\n    SemanticConfiguration, SemanticPrioritizedFields, SemanticField\n)\n\nindex = SearchIndex(\n    name=index_name,\n    fields=[\n        SearchField(name=\"id\", type=\"Edm.String\", key=True),\n        SearchField(name=\"content\", type=\"Edm.String\", searchable=True),\n        SearchField(name=\"embedding\", type=\"Collection(Edm.Single)\",\n                   vector_search_dimensions=3072,\n                   vector_search_profile_name=\"vector-profile\"),\n    ],\n    vector_search=VectorSearch(\n        profiles=[VectorSearchProfile(\n            name=\"vector-profile\",\n            algorithm_configuration_name=\"hnsw-algo\",\n            vectorizer_name=\"openai-vectorizer\"\n        )],\n        algorithms=[HnswAlgorithmConfiguration(name=\"hnsw-algo\")],\n        vectorizers=[AzureOpenAIVectorizer(\n            vectorizer_name=\"openai-vectorizer\",\n            parameters=AzureOpenAIVectorizerParameters(\n                resource_url=aoai_endpoint,\n                deployment_name=embedding_deployment,\n                model_name=embedding_model\n            )\n        )]\n    ),\n    semantic_search=SemanticSearch(\n        default_configuration_name=\"semantic-config\",\n        configurations=[SemanticConfiguration(\n            name=\"semantic-config\",\n            prioritized_fields=SemanticPrioritizedFields(\n                content_fields=[SemanticField(field_name=\"content\")]\n            )\n        )]\n    )\n)\n\nwith SearchIndexClient(endpoint, credential) as index_client:\n    index_client.create_or_update_index(index)\n```\n\n## Document Operations\n\n```python\nfrom azure.search.documents import SearchIndexingBufferedSender\n\n# Batch upload with automatic batching\nwith SearchIndexingBufferedSender(endpoint, index_name, credential) as sender:\n    sender.upload_documents(documents)\n\n# Direct operations via SearchClient\nwith SearchClient(endpoint, index_name, credential) as search_client:\n    search_client.upload_documents(documents)      # Add new\n    search_client.merge_documents(documents)       # Update existing\n    search_client.merge_or_upload_documents(documents)  # Upsert\n    search_client.delete_documents(documents)      # Remove\n```\n\n## Search Patterns\n\n```python\n# Basic search\nresults = search_client.search(search_text=\"query\")\n\n# Vector search\nfrom azure.search.documents.models import VectorizedQuery\n\nresults = search_client.search(\n    search_text=None,\n    vector_queries=[VectorizedQuery(\n        vector=embedding,\n        k_nearest_neighbors=5,\n        fields=\"embedding\"\n    )]\n)\n\n# Hybrid search (vector + keyword)\nresults = search_client.search(\n    search_text=\"query\",\n    vector_queries=[VectorizedQuery(vector=embedding, k_nearest_neighbors=5, fields=\"embedding\")],\n    query_type=\"semantic\",\n    semantic_configuration_name=\"semantic-config\"\n)\n\n# With filters\nresults = search_client.search(\n    search_text=\"query\",\n    filter=\"category eq 'technology'\",\n    select=[\"id\", \"title\", \"content\"],\n    top=10\n)\n```\n\n## Agentic Retrieval (Knowledge Bases)\n\nFor LLM-powered Q&A with answer synthesis, see [references\u002Fagentic-retrieval.md](references\u002Fagentic-retrieval.md).\n\nKey concepts:\n- **Knowledge Source**: Points to a search index\n- **Knowledge Base**: Wraps knowledge sources + LLM for query planning and synthesis\n- **Output modes**: `EXTRACTIVE_DATA` (raw chunks) or `ANSWER_SYNTHESIS` (LLM-generated answers)\n\n## Async Pattern\n\n```python\nfrom azure.search.documents.aio import SearchClient\n\nasync with SearchClient(endpoint, index_name, credential) as client:\n    results = await client.search(search_text=\"query\")\n    async for result in results:\n        print(result[\"title\"])\n```\n\n## Best Practices\n\n1. **Use environment variables** for endpoints, keys, and deployment names\n2. **Use `DefaultAzureCredential`** for code that runs locally (instead of API keys). Use a specific token credential for code that runs in Azure.\n3. **Use `SearchIndexingBufferedSender`** for batch uploads (handles batching\u002Fretries)\n4. **Always define semantic configuration** for agentic retrieval indexes\n5. **Use `create_or_update_index`** for idempotent index creation\n6. **Close clients** with context managers or explicit `close()`\n\n## Field Types Reference\n\n| EDM Type | Python | Notes |\n|----------|--------|-------|\n| `Edm.String` | str | Searchable text |\n| `Edm.Int32` | int | Integer |\n| `Edm.Int64` | int | Long integer |\n| `Edm.Double` | float | Floating point |\n| `Edm.Boolean` | bool | True\u002FFalse |\n| `Edm.DateTimeOffset` | datetime | ISO 8601 |\n| `Collection(Edm.Single)` | List[float] | Vector embeddings |\n| `Collection(Edm.String)` | List[str] | String arrays |\n\n## Error Handling\n\n```python\nfrom azure.core.exceptions import (\n    HttpResponseError,\n    ResourceNotFoundError,\n    ResourceExistsError\n)\n\ntry:\n    result = search_client.get_document(key=\"123\")\nexcept ResourceNotFoundError:\n    print(\"Document not found\")\nexcept HttpResponseError as e:\n    print(f\"Search error: {e.message}\")\n```\n",{"name":5,"license":15,"metadata":43,"description":46},{"author":12,"package":44,"version":45},"azure-search-documents","1.0.0","Azure AI Search SDK for Python. Use for vector search, hybrid search, semantic ranking, indexing, and skillsets.\nTriggers: \"azure-search-documents\", \"SearchClient\", \"SearchIndexClient\", \"vector search\", \"hybrid search\", \"semantic search\".\n","main","3f5b461b2314e01c400564051e92469114c5b68f",".github\u002Fplugins\u002Fazure-sdk-python\u002Fskills\u002Fazure-search-documents-py",368,"https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fskills","https:\u002F\u002Fmicrosoft.github.io\u002Fskills\u002F","https:\u002F\u002Fgithub.com\u002Fmicrosoft","https:\u002F\u002Fopensource.microsoft.com","Open source projects and samples from Microsoft","2026-11-14 06:48:25",{"source":8,"skillName":5,"commands":58,"prompt":68,"zipAvailable":69},[59,62,65],{"tool":60,"command":61},"npx","npx skills add microsoft\u002Fskills --skill azure-search-documents-py",{"tool":63,"command":64},"bunx","bunx skills add microsoft\u002Fskills --skill azure-search-documents-py",{"tool":66,"command":67},"pnpm","pnpm dlx skills add microsoft\u002Fskills --skill azure-search-documents-py","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fskills\u002Fblob\u002Fmain\u002F.github\u002Fplugins\u002Fazure-sdk-python\u002Fskills\u002Fazure-search-documents-py\u002FSKILL.md",true,[71,77,81,85,89,94,98,102],{"path":72,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":75,"sourceHtmlUrl":76},"examples\u002Farchitecture-examples.md","REFERENCE","参考文档",4968,"https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002F_collections\u002Frsmdt-the-startup\u002Fplugins\u002Fstart\u002Fskills\u002Farchitecture-design\u002Fexamples\u002Farchitecture-examples.md",{"path":78,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":79,"sourceHtmlUrl":80},"references\u002Fagentic-retrieval.md",7432,"https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fskills\u002Fblob\u002Fmain\u002F.github\u002Fplugins\u002Fazure-sdk-python\u002Fskills\u002Fazure-search-documents-py\u002Freferences\u002Fagentic-retrieval.md",{"path":82,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":83,"sourceHtmlUrl":84},"references\u002Fsemantic-ranking.md",7540,"https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fskills\u002Fblob\u002Fmain\u002F.github\u002Fplugins\u002Fazure-sdk-python\u002Fskills\u002Fazure-search-documents-py\u002Freferences\u002Fsemantic-ranking.md",{"path":86,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":87,"sourceHtmlUrl":88},"references\u002Fvector-search.md",6960,"https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fskills\u002Fblob\u002Fmain\u002F.github\u002Fplugins\u002Fazure-sdk-python\u002Fskills\u002Fazure-search-documents-py\u002Freferences\u002Fvector-search.md",{"path":90,"resourceType":91,"resourceTypeDesc":92,"sizeBytes":87,"sourceHtmlUrl":93},"scripts\u002Fsetup_agentic_retrieval.py","SCRIPT","脚本","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fskills\u002Fblob\u002Fmain\u002F.github\u002Fplugins\u002Fazure-sdk-python\u002Fskills\u002Fazure-search-documents-py\u002Fscripts\u002Fsetup_agentic_retrieval.py",{"path":95,"resourceType":91,"resourceTypeDesc":92,"sizeBytes":96,"sourceHtmlUrl":97},"scripts\u002Fsetup_vector_index.py",6143,"https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fskills\u002Fblob\u002Fmain\u002F.github\u002Fplugins\u002Fazure-sdk-python\u002Fskills\u002Fazure-search-documents-py\u002Fscripts\u002Fsetup_vector_index.py",{"path":99,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":100,"sourceHtmlUrl":101},"template.md",28628,"https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002F_collections\u002Frsmdt-the-startup\u002Fplugins\u002Fstart\u002Fskills\u002Farchitecture-design\u002Ftemplate.md",{"path":103,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":104,"sourceHtmlUrl":105},"validation.md",4312,"https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002F_collections\u002Frsmdt-the-startup\u002Fplugins\u002Fstart\u002Fskills\u002Farchitecture-design\u002Fvalidation.md",[107,115,142,179,206,226],{"skillNo":108,"name":109,"nameZh":6,"description":110,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":111,"categories":112,"keywords":113,"publishedAt":114},"01M741JBE1VH8A2Y3FXX2P7386","azure-advisor-cost-optimization","Analyze Azure Advisor Cost\u002FIdleResources findings and prevalidate deletion safety, including Service Groups. WHEN: \"find unused Azure resources\", \"what resources am I paying for but no longer using\", \"idle-resource savings\", \"historical idle-resource savings\", \"are these idle resources safe to delete\". Generic Cost, other subcategories, and cross-category requests use azure-advisor.","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fskills\u002Fblob\u002Fmain\u002F.github\u002Fplugins\u002Fazure-advisor\u002Fskills\u002Fazure-advisor-cost-optimization\u002FSKILL.md",[],[],"2026-11-11 15:11:14",{"skillNo":116,"name":117,"nameZh":6,"description":118,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":119,"categories":120,"keywords":126,"publishedAt":38},"01M69B4JCRQV5SPXAFCKXQRV0H","skill-creator","Guide for creating effective skills for AI coding agents working with Azure SDKs and Microsoft Foundry services. Use when creating new skills or updating existing skills.","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fskills\u002Fblob\u002Fmain\u002F.github\u002Fskills\u002Fskill-creator\u002FSKILL.md",[121],{"categoryNo":122,"categoryName":123,"taxonomyType":124,"taxonomyTypeDesc":125},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[127,128,129,130,9,20,131,132,133,134,26,32,135,136,137,138,139,140,141],"skill creator","creator","ai coding agents","coding agents","foundry","microsoft foundry","coding","guide","Existing","services","new","when","sdks","with","ai coding",{"skillNo":143,"name":144,"nameZh":6,"description":145,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":146,"categories":147,"keywords":158,"publishedAt":38},"01M69B4FD57V1MYKRWAJJECTZ7","podcast-generation","Generate AI-powered podcast-style audio narratives using Azure OpenAI's GPT Realtime Mini model via WebSocket. Use when building text-to-speech features, audio narrative generation, podcast creation from content, or integrating with Azure OpenAI Realtime API for real audio output. Covers full-stack implementation from React frontend to Python FastAPI backend with WebSocket streaming.","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fskills\u002Fblob\u002Fmain\u002F.github\u002Fskills\u002Fpodcast-generation\u002FSKILL.md",[148,151,154,157],{"categoryNo":149,"categoryName":150,"taxonomyType":124,"taxonomyTypeDesc":125},"01JBSEEDCAT000000000000009","前端开发",{"categoryNo":152,"categoryName":153,"taxonomyType":124,"taxonomyTypeDesc":125},"01JBSEEDCAT000000000000010","后端与接口",{"categoryNo":155,"categoryName":156,"taxonomyType":124,"taxonomyTypeDesc":125},"01JBSEEDCAT000000000000016","图像与多媒体",{"categoryNo":122,"categoryName":123,"taxonomyType":124,"taxonomyTypeDesc":125},[159,160,161,162,20,163,164,165,33,166,167,168,169,170,171,138,172,173,174,175,176,177,178],"podcast generation","generation","podcast","frontend","azure openai","realtime","react","Audio","creation","model","fastapi backend","python fastapi","fastapi","api","streaming","text","narrative","stack","generate","features",{"skillNo":180,"name":181,"nameZh":6,"description":182,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":183,"categories":184,"keywords":186,"publishedAt":38},"01M69B4CYPM5SESY5ZWEFDXEQZ","microsoft-docs","Understand Microsoft technologies by querying official documentation. Use whenever the user asks how something works, wants tutorials, needs configuration options, limits, quotas, or best practices for any Microsoft technology (Azure, .NET, M365, Windows, Power Platform, etc.)—even if they don't mention \"docs.\" If the question is about understanding a concept rather than writing code, this is the right skill.","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fskills\u002Fblob\u002Fmain\u002F.github\u002Fskills\u002Fmicrosoft-docs\u002FSKILL.md",[185],{"categoryNo":122,"categoryName":123,"taxonomyType":124,"taxonomyTypeDesc":125},[187,9,188,189,190,191,192,20,193,194,195,196,197,198,199,200,26,201,60,202,32,203,204,205],"microsoft docs","docs","documentation","windows","official","cli","MCP","writing","power platform","mcp server","back","concept","power","platform","technology","question","learn mcp","learn","best practices",{"skillNo":207,"name":208,"nameZh":6,"description":209,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":210,"categories":211,"keywords":213,"publishedAt":38},"01M69B49C3YPD1ZJ3NC0HPCY58","mcp-builder","Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP), Node\u002FTypeScript (MCP SDK), or C#\u002F.NET (Microsoft MCP SDK).","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fskills\u002Fblob\u002Fmain\u002F.github\u002Fskills\u002Fmcp-builder\u002FSKILL.md",[212],{"categoryNo":122,"categoryName":123,"taxonomyType":124,"taxonomyTypeDesc":125},[214,193,215,216,37,9,33,217,218,168,219,138,220,221,222,134,223,224,26,225,32,140],"mcp builder","builder","llms","typescript","model context protocol","building mcp servers","node","integrate","Tools","context","fastmcp","mcp sdk",{"skillNo":227,"name":228,"nameZh":6,"description":229,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":230,"categories":231,"keywords":244,"publishedAt":38},"01M69B467A7CDMFQ0T6Q0ZA12Q","kql","KQL language expertise for writing correct, efficient Kusto Query Language queries. Covers syntax gotchas, join patterns, dynamic types, datetime pitfalls, regex patterns, serialization, memory management, result-size discipline, and advanced functions (geo, vector, graph). USE THIS SKILL whenever writing, debugging, or reviewing KQL queries — even simple ones — because the gotchas section prevents the most common errors that waste tool calls and cause expensive retry cascades. Trigger on: KQL, Kusto, ADX, Azure Data Explorer, Fabric Real-Time Intelligence, EventHouse, Log Analytics, log analysis, data exploration, time series, anomaly detection, summarize, where clause, join, extend, project, let statement, parse operator, extract function, any mention of pipe-forward query syntax.","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fskills\u002Fblob\u002Fmain\u002F.github\u002Fskills\u002Fkql\u002FSKILL.md",[232,235,236,241],{"categoryNo":233,"categoryName":234,"taxonomyType":124,"taxonomyTypeDesc":125},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":122,"categoryName":123,"taxonomyType":124,"taxonomyTypeDesc":125},{"categoryNo":237,"categoryName":238,"taxonomyType":239,"taxonomyTypeDesc":240},"01JBSEEDCAT000000000000033","数据分析师","PROFESSION","适用职业",{"categoryNo":242,"categoryName":243,"taxonomyType":239,"taxonomyTypeDesc":240},"01JBSEEDCAT000000000000036","内容创作者",[228,20,194,245,246,247,248,249,250,251,34,252,253,254,255,256,257,258,259,26,260,261,262],"analytics","memory","summarize","exploration","parse","project","time","join","log","pipe","datetime","discipline","time series","regex","anomaly","section","tool","geo"]