{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"github-microsoft--generative-ai-for-beginners","slug":"microsoft--generative-ai-for-beginners","name":"generative-ai-for-beginners","type":"repo","url":"https://github.com/microsoft/generative-ai-for-beginners","page_url":"https://unfragile.ai/microsoft--generative-ai-for-beginners","categories":["prompt-engineering"],"tags":["ai","azure","chatgpt","dall-e","generative-ai","generativeai","gpt","language-model","llms","microsoft-for-beginners","openai","prompt-engineering","semantic-search","transformers"],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"github-microsoft--generative-ai-for-beginners__cap_0","uri":"capability://text.generation.language.structured.llm.fundamentals.curriculum.delivery","name":"structured-llm-fundamentals-curriculum-delivery","description":"Delivers a 21-lesson progressive curriculum structured as 'Learn' (conceptual) and 'Build' (hands-on) modules that scaffold from LLM basics through advanced applications. Uses a modular Jupyter Notebook architecture with embedded code examples in both Python and TypeScript, allowing learners to execute concepts immediately within their development environment rather than reading static documentation.","intents":["I need to understand how LLMs work from first principles before building with them","I want a structured path from generative AI fundamentals to building production applications","I need to learn in both Python and TypeScript without switching between separate courses"],"best_for":["developers new to generative AI seeking structured onboarding","teams building internal AI literacy programs","students in academic settings needing comprehensive GenAI foundations"],"limitations":["Curriculum is fixed and linear — no adaptive learning paths based on learner background","Jupyter Notebook format requires local runtime setup; not accessible in browser without additional hosting","Lessons are snapshot-in-time; LLM landscape evolves faster than course updates typically occur"],"requires":["Python 3.8+ or Node.js 14+ for code execution","Jupyter Notebook environment or VS Code with Jupyter extension","API keys for OpenAI, Azure OpenAI, or compatible LLM providers","Git for cloning the repository"],"input_types":["markdown lesson content","executable Python/TypeScript code cells","API documentation references"],"output_types":["executed notebook outputs","generated text/images from API calls","learner-created application code"],"categories":["text-generation-language","educational-curriculum"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-microsoft--generative-ai-for-beginners__cap_1","uri":"capability://planning.reasoning.prompt.engineering.technique.progression","name":"prompt-engineering-technique-progression","description":"Teaches prompt engineering through a two-tier approach: foundational techniques (clarity, specificity, role-based prompting) in Lesson 4, then advanced techniques (chain-of-thought, few-shot examples, system prompts) in Lesson 5. Each technique is demonstrated with concrete examples and code snippets showing how to structure prompts for OpenAI and Azure OpenAI APIs, with measurable improvements in output quality shown through side-by-side comparisons.","intents":["I need to understand why my LLM outputs are poor quality and how to fix them","I want to learn specific prompt patterns that work reliably across different LLM providers","I need to teach my team effective prompting without them having to experiment blindly"],"best_for":["developers building LLM-powered features who want to optimize output quality","product teams designing AI-assisted workflows","non-technical stakeholders learning to interact effectively with LLMs"],"limitations":["Prompt engineering is empirical and model-dependent — techniques that work for GPT-4 may not transfer equally to open-source models or older API versions","No automated prompt optimization or testing framework provided — learners must manually iterate","Curriculum focuses on English prompting; multilingual prompt engineering patterns not deeply covered"],"requires":["OpenAI API key or Azure OpenAI access","Understanding of basic LLM concepts (covered in Lesson 1)","Ability to read and modify JSON/Python code for API calls"],"input_types":["natural language prompts","structured prompt templates","example outputs for comparison"],"output_types":["improved LLM responses","prompt templates for reuse","quality metrics showing improvement"],"categories":["planning-reasoning","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-microsoft--generative-ai-for-beginners__cap_10","uri":"capability://automation.workflow.low.code.ai.application.development.with.azure.ai.studio","name":"low-code-ai-application-development-with-azure-ai-studio","description":"Lesson 10 teaches building AI applications using Azure AI Studio, a low-code/no-code platform that abstracts away API management and code complexity. Provides guided workflows for creating chat applications, search applications, and function-calling agents without writing code. Demonstrates how to configure models, define prompts, test interactions, and deploy applications through a visual interface. Enables non-technical users and rapid prototypers to build functional AI applications without software development expertise.","intents":["I need to build an AI application quickly without writing code","I want to test different prompts and configurations without deploying code changes","I need a visual interface to manage AI application configuration and deployment"],"best_for":["non-technical business users building AI applications","rapid prototypers validating AI product ideas","teams in Microsoft Azure ecosystems"],"limitations":["Azure AI Studio is Microsoft-specific; no equivalent coverage for other cloud providers (AWS, GCP)","Low-code approach limits customization for complex requirements","No guidance on migrating from low-code prototypes to production code","Pricing and cost management not deeply covered; users may incur unexpected costs"],"requires":["Azure account with appropriate permissions","Access to Azure AI Studio (may require specific subscription tier)","Basic understanding of AI concepts (Lessons 1-3)"],"input_types":["visual configuration","prompt text","model selection"],"output_types":["deployed AI application","chat interface","API endpoint"],"categories":["automation-workflow","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-microsoft--generative-ai-for-beginners__cap_11","uri":"capability://planning.reasoning.llm.model.comparison.and.selection.framework","name":"llm-model-comparison-and-selection-framework","description":"Lesson 2 teaches systematic model selection by comparing different LLMs (GPT-4, GPT-3.5, open-source models) across dimensions: cost, latency, quality, context window, and specialized capabilities. Provides a decision framework for choosing models based on use case requirements, with guidance on trade-offs between proprietary and open-source, larger and smaller models. Explains how to evaluate models empirically by testing on representative tasks rather than relying on marketing claims.","intents":["I need to choose the right LLM for my use case without trying every option","I want to understand the trade-offs between cost, quality, and latency for different models","I need to evaluate whether a smaller open-source model can work instead of expensive proprietary APIs"],"best_for":["architects designing AI systems and selecting technology","teams optimizing for cost or latency constraints","organizations evaluating build-vs-buy decisions for LLM capabilities"],"limitations":["Model landscape evolves rapidly; curriculum becomes outdated as new models are released","Comparison is qualitative rather than quantitative; no standardized benchmarks provided","No guidance on running benchmark tests to compare models on your specific task","Specialized models (domain-specific, multimodal) not comprehensively covered"],"requires":["Understanding of LLM fundamentals (Lesson 1)","Access to multiple LLM APIs for testing (OpenAI, Azure, Hugging Face, etc.)","Representative test cases for your use case"],"input_types":["use case requirements","test prompts","performance criteria"],"output_types":["model comparison matrix","selection recommendation","cost/quality trade-off analysis"],"categories":["planning-reasoning","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-microsoft--generative-ai-for-beginners__cap_12","uri":"capability://text.generation.language.multilingual.curriculum.delivery.and.localization","name":"multilingual-curriculum-delivery-and-localization","description":"The curriculum is available in multiple languages (Chinese, Spanish, Portuguese, Japanese) with translations of all lessons and code examples. Each translation is maintained in the repository with language-specific directories, enabling learners to access the full course in their native language. Demonstrates commitment to global accessibility and removes language barriers for non-English speakers learning generative AI.","intents":["I need to learn generative AI in my native language, not English","I want to teach generative AI to teams in non-English-speaking regions","I need culturally appropriate examples and explanations for my context"],"best_for":["non-English-speaking developers and students","organizations building AI literacy in international teams","educational institutions in non-English-speaking countries"],"limitations":["Translations may lag behind English version; new lessons may not be immediately available in all languages","Code examples and API documentation are still primarily in English","Cultural adaptation is limited; examples may not reflect local contexts or use cases","Translation quality varies; some languages may have more complete translations than others"],"requires":["Ability to read the target language","Access to the translated course materials in the repository"],"input_types":["course content in target language","code examples"],"output_types":["localized learning materials","translated code examples"],"categories":["text-generation-language","educational-curriculum"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-microsoft--generative-ai-for-beginners__cap_2","uri":"capability://safety.moderation.responsible.ai.and.ethical.guidelines.framework","name":"responsible-ai-and-ethical-guidelines-framework","description":"Provides a structured framework for responsible AI development covering bias detection, fairness assessment, transparency, and ethical considerations specific to generative AI. Lesson 3 integrates responsible AI practices as a foundational concept rather than an afterthought, with guidance on identifying potential harms, testing for bias in model outputs, and implementing safeguards. Uses Microsoft's responsible AI principles as the pedagogical framework.","intents":["I need to understand potential harms my AI application could cause before deploying it","I want to audit my LLM outputs for bias and fairness issues","I need to explain to stakeholders why responsible AI practices matter in our product"],"best_for":["teams building production AI systems who need compliance frameworks","organizations establishing AI governance policies","developers in regulated industries (finance, healthcare, government)"],"limitations":["Framework is principles-based rather than providing automated testing tools — implementation requires manual review and custom testing","Responsible AI is context-dependent; guidelines are general and may not address domain-specific ethical concerns","No integration with automated bias detection tools or model cards — learners must implement their own evaluation pipelines"],"requires":["Understanding of basic LLM capabilities and limitations (Lesson 1)","Familiarity with the specific use case and potential stakeholder impacts","Access to model outputs for manual evaluation"],"input_types":["LLM outputs","user demographics and context","ethical concern scenarios"],"output_types":["bias assessment reports","fairness evaluation results","mitigation strategy recommendations"],"categories":["safety-moderation","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-microsoft--generative-ai-for-beginners__cap_3","uri":"capability://code.generation.editing.multi.application.type.hands.on.building","name":"multi-application-type-hands-on-building","description":"Provides executable code examples and architectural patterns for building six distinct types of generative AI applications: text generation (Lesson 6), chat/conversational (Lesson 7), semantic search (Lesson 8), image generation (Lesson 9), low-code/no-code (Lesson 10), and function-calling-integrated (Lesson 11). Each lesson includes working code in Python and TypeScript that connects to actual APIs (OpenAI, Azure OpenAI, DALL-E), allowing learners to build and deploy functional applications rather than just understanding concepts.","intents":["I need to build a chatbot and want to see working code I can adapt for my use case","I want to understand the architectural differences between text generation, chat, and search applications","I need to integrate LLM capabilities with external tools and APIs in my application"],"best_for":["full-stack developers building AI-powered features","startup founders prototyping AI products quickly","teams evaluating which application architecture fits their use case"],"limitations":["Code examples are simplified for learning; production deployments require additional error handling, rate limiting, and cost management","Examples use cloud APIs (OpenAI, Azure) — no guidance on deploying with open-source models or on-premises LLMs","Low-code lesson (Lesson 10) covers Azure AI Studio but is specific to Microsoft ecosystem; other platforms not covered","Function calling examples are OpenAI-specific; Anthropic and other providers' function-calling patterns differ"],"requires":["Python 3.8+ or Node.js 14+","API keys for OpenAI, Azure OpenAI, or DALL-E","Basic understanding of REST APIs and async/await patterns","Jupyter Notebook or VS Code environment"],"input_types":["user prompts/queries","conversation history","search queries","image generation descriptions","function definitions"],"output_types":["generated text","conversation responses","ranked search results","generated images","function call results"],"categories":["code-generation-editing","text-generation-language","image-visual","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-microsoft--generative-ai-for-beginners__cap_4","uri":"capability://memory.knowledge.semantic.search.and.rag.architecture.teaching","name":"semantic-search-and-rag-architecture-teaching","description":"Lesson 8 teaches semantic search by explaining vector embeddings, similarity matching, and retrieval-augmented generation (RAG) concepts, then provides code examples showing how to embed documents, store them in vector databases, and retrieve relevant context to augment LLM prompts. Lesson 13 (Advanced Topics) goes deeper into RAG patterns, vector database selection, and chunking strategies. The curriculum explains the architectural flow: documents → embeddings → vector store → retrieval → LLM context augmentation.","intents":["I need to build a search feature that understands meaning, not just keyword matching","I want to augment my LLM with domain-specific knowledge without fine-tuning","I need to understand how to structure documents and embeddings for efficient retrieval"],"best_for":["developers building knowledge-base-powered chatbots or Q&A systems","teams implementing document search with semantic understanding","organizations building domain-specific AI assistants"],"limitations":["Vector database selection and setup is not deeply covered — learners must evaluate options (Pinecone, Weaviate, Milvus) independently","Chunking strategies are mentioned but not comprehensively taught; optimal chunk size is use-case-dependent","No guidance on handling document updates, versioning, or maintaining consistency between source documents and embeddings","Embedding costs and latency trade-offs not quantified; learners may not understand performance implications"],"requires":["Understanding of embeddings and vector similarity (covered in Lesson 8)","Access to embedding API (OpenAI, Azure OpenAI, or open-source models)","Vector database setup (local or cloud-hosted)","Document corpus to embed and search"],"input_types":["documents or text passages","user queries","embedding vectors"],"output_types":["embedding vectors","ranked relevant documents","augmented LLM prompts with context"],"categories":["memory-knowledge","search-retrieval","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-microsoft--generative-ai-for-beginners__cap_5","uri":"capability://code.generation.editing.open.source.and.fine.tuning.model.alternatives","name":"open-source-and-fine-tuning-model-alternatives","description":"Lesson 14 (Advanced Topics) covers open-source LLM alternatives (Hugging Face models, Llama, Mistral) and their trade-offs versus proprietary APIs. Lesson 15 teaches fine-tuning approaches: parameter-efficient methods (LoRA, QLoRA) and full fine-tuning, with guidance on when each is appropriate. Provides code examples showing how to load open-source models, prepare training data, and execute fine-tuning workflows using libraries like Hugging Face Transformers and PEFT.","intents":["I need to reduce API costs by running models locally or on my own infrastructure","I want to customize a model for my specific domain or task without building from scratch","I need to understand the trade-offs between fine-tuning, RAG, and prompt engineering for knowledge injection"],"best_for":["teams with budget constraints or data privacy requirements","organizations building domain-specific AI systems","developers optimizing for latency or cost at scale"],"limitations":["Fine-tuning requires significant computational resources (GPU/TPU) and ML expertise — not accessible to all learners","Open-source model quality varies widely; curriculum doesn't provide systematic comparison framework","LoRA/QLoRA are covered conceptually but require deep understanding of model architecture to debug","No guidance on model licensing, compliance, or commercial use restrictions for open-source models"],"requires":["Python 3.8+","GPU access for fine-tuning (NVIDIA CUDA or equivalent)","Hugging Face Transformers and PEFT libraries","Training dataset in appropriate format","Understanding of LLM fundamentals (Lessons 1-2)"],"input_types":["pre-trained model weights","training data (text, instruction-response pairs)","fine-tuning hyperparameters"],"output_types":["fine-tuned model weights","adapter weights (LoRA)","model performance metrics"],"categories":["code-generation-editing","data-processing-analysis","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-microsoft--generative-ai-for-beginners__cap_6","uri":"capability://planning.reasoning.ux.design.patterns.for.ai.applications","name":"ux-design-patterns-for-ai-applications","description":"Lesson 12 teaches UX design principles specific to AI applications: handling uncertainty and hallucinations, designing for transparency, managing user expectations, and providing feedback mechanisms. Covers patterns like confidence scores, source attribution, fallback responses, and progressive disclosure of AI limitations. Provides design guidance rather than code, focusing on how to structure user interactions to account for LLM unreliability and the need for human oversight.","intents":["I need to design an interface that helps users understand when to trust AI outputs and when to verify them","I want to surface model confidence or uncertainty to users appropriately","I need to design workflows that incorporate human review or correction of AI outputs"],"best_for":["product designers building AI-powered features","UX researchers evaluating AI application usability","teams designing customer-facing AI products"],"limitations":["Lesson is design-focused without interactive prototypes or usability testing data","No quantitative guidance on when to show confidence scores vs hide them","Patterns are general; domain-specific UX considerations (medical, legal, financial) not deeply covered","No integration with actual LLM APIs to demonstrate real-time uncertainty handling"],"requires":["Understanding of basic UX design principles","Familiarity with LLM capabilities and limitations (Lessons 1-3)","Design tools (Figma, Adobe XD, etc.) for prototyping"],"input_types":["LLM output with confidence metadata","user interaction patterns","error scenarios"],"output_types":["UI mockups","interaction patterns","design guidelines"],"categories":["planning-reasoning","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-microsoft--generative-ai-for-beginners__cap_7","uri":"capability://safety.moderation.ai.application.security.and.threat.modeling","name":"ai-application-security-and-threat-modeling","description":"Lesson 16 (Advanced Topics) covers security considerations for AI applications: prompt injection attacks, data privacy in API calls, model poisoning, and securing API keys. Provides threat modeling guidance specific to generative AI systems, explaining attack vectors like adversarial prompts designed to bypass safety guidelines, and mitigation strategies like input validation, rate limiting, and secure credential management. Emphasizes that security in AI applications requires both traditional software security (API key management) and AI-specific concerns (prompt injection).","intents":["I need to secure my AI application against prompt injection and adversarial attacks","I want to understand data privacy implications of sending user data to LLM APIs","I need to implement security controls for production AI systems"],"best_for":["security engineers evaluating AI application risks","teams building production AI systems in regulated industries","developers implementing security controls for LLM integrations"],"limitations":["Prompt injection defenses are evolving; curriculum covers known patterns but new attack vectors emerge regularly","No automated security testing tools or frameworks provided — learners must implement custom validation","Guidance is general; domain-specific security requirements (HIPAA, PCI-DSS, SOC 2) not covered","No integration with security scanning tools or vulnerability databases"],"requires":["Understanding of basic application security (authentication, authorization, encryption)","Familiarity with LLM capabilities and limitations (Lessons 1-3)","Knowledge of the specific regulatory requirements for your domain"],"input_types":["user prompts","API requests","model outputs"],"output_types":["threat models","security recommendations","validation rules"],"categories":["safety-moderation","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-microsoft--generative-ai-for-beginners__cap_8","uri":"capability://tool.use.integration.function.calling.and.tool.integration.patterns","name":"function-calling-and-tool-integration-patterns","description":"Lesson 11 teaches function calling (also called tool use) as a pattern for extending LLM capabilities by defining external functions the model can invoke. Provides code examples showing how to define function schemas, handle model-generated function calls, and execute them, using OpenAI's function calling API as the primary example. Explains the architectural pattern: user query → LLM generates function call → application executes function → result fed back to LLM → final response. Enables building agents that can interact with external systems, APIs, and databases.","intents":["I need my LLM to call external APIs or databases to answer user queries","I want to build an agent that can take actions in external systems based on user requests","I need to structure function definitions so the LLM can reliably call the right function with correct parameters"],"best_for":["developers building LLM agents with external tool access","teams implementing AI assistants that interact with business systems","builders creating autonomous workflows triggered by natural language"],"limitations":["Function calling is provider-specific; OpenAI, Anthropic, and others have different schemas and behaviors","No automatic parameter validation — LLM may generate invalid function calls that require error handling","Curriculum focuses on synchronous function calls; async/long-running operations not deeply covered","No guidance on managing function call costs or rate limiting in production"],"requires":["OpenAI API key or compatible provider","Understanding of JSON schema for function definitions","Ability to implement and expose functions in your application","Error handling for invalid function calls"],"input_types":["user queries","function definitions (JSON schema)","function execution results"],"output_types":["function calls (structured JSON)","function results","final LLM response"],"categories":["tool-use-integration","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-microsoft--generative-ai-for-beginners__cap_9","uri":"capability://image.visual.image.generation.and.multimodal.application.building","name":"image-generation-and-multimodal-application-building","description":"Lesson 9 teaches image generation using DALL-E API, covering prompt engineering for image generation (different from text prompts), image editing and variation capabilities, and integration into applications. Provides code examples showing how to call the DALL-E API, handle generated images, and build workflows that combine text and image generation. Explains the differences between image generation, editing, and variation endpoints, and when to use each.","intents":["I need to generate images programmatically based on user descriptions","I want to build an application that combines text and image generation","I need to understand how to prompt for images effectively (different from text prompting)"],"best_for":["developers building creative AI applications (design tools, content creation)","teams implementing multimodal AI features","product builders exploring image generation capabilities"],"limitations":["DALL-E is the primary example; other image generation models (Stable Diffusion, Midjourney) not covered","Image generation costs are significant; curriculum doesn't provide cost optimization strategies","No guidance on copyright, licensing, or ethical use of generated images","Image editing and variation capabilities are limited compared to dedicated image editing tools"],"requires":["OpenAI API key with DALL-E access","Understanding of image generation prompt engineering (different from text prompting)","Ability to handle image URLs and storage"],"input_types":["text descriptions (prompts)","existing images (for editing/variation)","image parameters (size, quality)"],"output_types":["generated images (URLs)","edited images","image variations"],"categories":["image-visual","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":56,"verified":false,"data_access_risk":"high","permissions":["Python 3.8+ or Node.js 14+ for code execution","Jupyter Notebook environment or VS Code with Jupyter extension","API keys for OpenAI, Azure OpenAI, or compatible LLM providers","Git for cloning the repository","OpenAI API key or Azure OpenAI access","Understanding of basic LLM concepts (covered in Lesson 1)","Ability to read and modify JSON/Python code for API calls","Azure account with appropriate permissions","Access to Azure AI Studio (may require specific subscription tier)","Basic understanding of AI concepts (Lessons 1-3)"],"failure_modes":["Curriculum is fixed and linear — no adaptive learning paths based on learner background","Jupyter Notebook format requires local runtime setup; not accessible in browser without additional hosting","Lessons are snapshot-in-time; LLM landscape evolves faster than course updates typically occur","Prompt engineering is empirical and model-dependent — techniques that work for GPT-4 may not transfer equally to open-source models or older API versions","No automated prompt optimization or testing framework provided — learners must manually iterate","Curriculum focuses on English prompting; multilingual prompt engineering patterns not deeply covered","Azure AI Studio is Microsoft-specific; no equivalent coverage for other cloud providers (AWS, GCP)","Low-code approach limits customization for complex requirements","No guidance on migrating from low-code prototypes to production code","Pricing and cost management not deeply covered; users may incur unexpected costs","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.9578315595807687,"quality":0.35,"ecosystem":0.6000000000000001,"match_graph":0.25,"freshness":0.75,"weights":{"adoption":0.3,"quality":0.2,"ecosystem":0.15,"match_graph":0.3,"freshness":0.05}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"active","updated_at":"2026-05-24T12:16:22.062Z","last_scraped_at":"2026-05-03T13:58:21.997Z","last_commit":"2026-04-30T09:47:28Z"},"community":{"stars":110137,"forks":59074,"weekly_downloads":null,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=microsoft--generative-ai-for-beginners","compare_url":"https://unfragile.ai/compare?artifact=microsoft--generative-ai-for-beginners"}},"signature":"+zt3yLE2aEDmFDzb3YKu9s7dY3sRgtp5+75wE/SOyLbnCMtkZ1yXNjDyDngDGtr60QYoPTGG/jBNfn2He0tEBw==","signedAt":"2026-06-20T13:22:47.845Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/microsoft--generative-ai-for-beginners","artifact":"https://unfragile.ai/microsoft--generative-ai-for-beginners","verify":"https://unfragile.ai/api/v1/verify?slug=microsoft--generative-ai-for-beginners","publicKey":"https://unfragile.ai/api/v1/trust-passport-public-key","spec":"https://unfragile.ai/trust","schema":"https://unfragile.ai/schema.json","docs":"https://unfragile.ai/docs"}}