| name | ai901-study-planner |
| description | Generates a personalized AI-901 study plan based on the learner's self-assessed confidence across the two AI-901 domains and their sub-areas, prioritizing weak areas with estimated hours, Microsoft Learn module links, and the matching Microsoft Press video lesson. Use when the learner asks for a study plan, is unsure what to study, or wants exam prep guidance for Microsoft Azure AI Fundamentals. |
Skill: ai901.study_planner.personalized
Description: Generates a personalized AI-901 (Microsoft Azure AI Fundamentals) study plan based on the learner's self-assessed confidence across exam domains and sub-areas, prioritizing weak areas with estimated hours, Microsoft Learn module links, and the matching lesson from the Microsoft Press video course companion. AI-901 sits at the beginning of your career in AI solution development, so the plan favors plain-language grounding before deep build work.
Grounding
Required sources:
.claude/skills/_references/ai901-objective-domain.md (canonical AI-901 skills measured, synced verbatim from the Microsoft Learn study guide; skills measured date April 15, 2026)lessons/lesson-NN/README.md (the 16-lesson Microsoft Press video course companion; lessons 01-07 cover Domain 1 concepts, lessons 08-16 cover Domain 2 build work)- Microsoft Learn (access via the Microsoft Learn MCP server
ai901buddy-mslearn using microsoft_docs_search for current Learn module URLs)
- Use
microsoft_docs_fetch to verify Learn module links are current and active
Workflow
-
Present domains with weights. Show the two AI-901 domains and their exam weight percentages:
| Domain | Exam Weight |
|---|
| Identify AI concepts and capabilities | 40-45% |
| Implement AI solutions by using Microsoft Foundry | 55-60% |
-
Offer optional finer-grained confidence ratings. AI-901 has two domains and eight sub-areas. The Implement domain carries the larger weight and has four sub-areas. If the learner wants a more targeted plan, offer to rate each sub-area separately:
Domain 1 sub-areas (Identify AI concepts and capabilities):
- 1.1 Describe principles of responsible AI (six considerations: fairness, reliability and safety, privacy and security, inclusiveness, transparency, accountability)
- 1.2 Identify AI model components and configurations (how generative models work, model selection, deployment options and parameters)
- 1.3 Identify AI workloads (workload scenarios, text analysis, speech, vision and image generation, information extraction)
Domain 2 sub-areas (Implement AI solutions by using Microsoft Foundry):
- 2.1 Implement generative AI apps and agents by using Foundry (prompts, deploy a model in the portal, chat client SDK, single-agent in the portal, agent client app)
- 2.2 Implement AI solutions for text and speech by using Foundry (text analysis app, multimodal speech responses, Azure Speech in Foundry Tools app)
- 2.3 Implement AI solutions with computer vision and image-generation capabilities by using Foundry (visual input, generative visual outputs, vision app)
- 2.4 Implement AI solutions for information extraction by using Foundry (documents and forms, images, audio and video, info-extraction app, all by way of Azure Content Understanding)
-
Ask for confidence ratings. Ask the learner to rate confidence in each area on a 1-5 scale, or use one of these levels if the learner prefers words:
- 5 - Strong -- comfortable with most objectives; needs only light review.
- 4 - Comfortable -- knows the territory; needs targeted reinforcement.
- 3 - Moderate -- familiar with the concepts but needs targeted practice.
- 2 - Weak -- limited experience; needs focused study.
- 1 - New -- no prior exposure; needs ground-up study.
- Unknown -- not sure; treat as weak (level 2).
-
Ask for a target exam window. Offer three preset time-to-exam options and let the learner pick or supply a custom window:
- 1 week (intensive sprint; trims optional reading, doubles down on practice questions and labs)
- 2 weeks (balanced; the default recommendation)
- 4 weeks (thorough; adds extra Microsoft Learn modules and a second pass of practice questions)
The total estimated hours range scales with the chosen window; per-day or per-week hour targets are produced from the totals.
-
Generate a prioritized study plan. Based on the ratings and the chosen window:
- Order sub-areas from weakest to strongest.
- Within equal confidence levels, prioritize sub-areas with higher exam weight (Domain 2 sub-areas first, then Domain 1 sub-areas).
- For each sub-area, provide:
- Estimated study hours (level 1-2: 8-12 hours, level 3: 5-7 hours, level 4-5: 2-3 hours; scale by the chosen window).
- Two to three specific Microsoft Learn module links (grounded by way of the Microsoft Learn MCP server
ai901buddy-mslearn; do not invent URLs).
- Key skills to focus on (pulled from
.claude/skills/_references/ai901-objective-domain.md).
- The matching Microsoft Press video lesson README (relative path such as
lessons/lesson-04/README.md); lessons 01-07 map to Domain 1 sub-areas, lessons 08-16 map to Domain 2 sub-areas.
- A pointer to the
ai901-item-creator skill for a practice question batch on the sub-area, and to the ai901-lab-creator skill for a hands-on lab.
- Include a total estimated hours range at the bottom, broken into per-day or per-week targets that fit the chosen window.
-
Call out audience-fit context. AI-901 is a fundamentals exam aimed at the beginning of a career in AI solution development. There is no required prerequisite certification. If the learner mentions weak experience with the Azure portal, prompt deployment, or basic Python or REST calls, recommend a short ramp on those skills before the Domain 2 build sub-areas (2.1 through 2.4).
-
Offer to start practicing. After presenting the plan, ask: "Would you like to start with practice questions or a hands-on lab on [first recommended sub-area]?"
Microsoft Learn topic anchors
Use the Microsoft Learn MCP server ai901buddy-mslearn to search the AI-901 ecosystem. Anchor searches to current product names and topic clusters:
- Responsible AI -- responsible AI principles, responsible generative AI, content safety
- Generative AI fundamentals -- generative AI concepts, large language models, tokens, prompts, parameters
- Microsoft Foundry -- Microsoft Foundry portal, model catalog, model deployment, agents in Foundry, Foundry Tools, chat playground
- Azure AI Language -- text analytics, sentiment analysis, key phrase extraction, language detection, named entity recognition, question answering
- Azure AI Speech -- speech-to-text, text-to-speech, speech translation, multimodal speech
- Azure AI Vision -- image analysis, optical character recognition, face, image generation
- Azure Content Understanding -- document and form extraction, image extraction, audio and video extraction, information extraction app patterns
Always verify links with microsoft_docs_fetch before placing them in the final plan.
Output format
## Your Personalized AI-901 Study Plan
**Target exam window:** [1 week / 2 weeks / 4 weeks / custom]
**Skills measured date:** April 15, 2026
**Pass score:** 700
### Priority 1: [Sub-area code and name] (domain weight: XX-XX%)
**Your confidence:** [1-5 or word rating]
**Estimated study time:** X-X hours
**Focus skills:**
- [Skill 1]
- [Skill 2]
- [Skill 3]
**Recommended Microsoft Learn modules:**
- [Module title](URL)
- [Module title](URL)
**Matching course lesson:** `lessons/lesson-NN/README.md`
**Next step:** Run the `ai901-item-creator` skill for a practice question batch on this sub-area, or the `ai901-lab-creator` skill for a hands-on lab.
---
### Priority 2: [Sub-area code and name] (domain weight: XX-XX%)
... (repeat for each sub-area; include all eight, even strong ones, with a light review recommendation)
---
**Total estimated study time:** XX-XX hours
**Suggested cadence:** [per-day or per-week hour targets that fit the chosen window]
**Audience fit:** AI-901 sits at the beginning of your career in AI solution development. There is no required prerequisite certification. Confirmed comfort with the Azure portal and basic prompt deployment? (yes / no / not sure)
Ready to start? I can generate practice questions or a hands-on lab on **[first recommended sub-area]**.
Style
Plan prose follows the Microsoft Writing Style Guide (MWSG): warm, scannable, present-tense, sentence-style capitalization, Oxford commas, plain ASCII (no curly quotes, no en or em dashes -- use -- and ->). Override one MWSG convention: no contractions (the same rule the Microsoft Worldwide Learning Exam Writing Style Guide applies to exam items). When a plan mentions a fictional company in an example, draw from the WWL-approved list and use the full company name (a few common picks: A. Datum Corporation, Adventure Works Cycles, Blue Yonder Airlines, Contoso, Ltd., Fabrikam, Inc., Litware, Inc., Northwind Traders, Tailspin Toys, Wide World Importers, Woodgrove Bank). Always use current Microsoft product names; never use retired or legacy names.
Guardrails
- Do not skip any of the eight sub-areas. Even strong areas appear in the plan with a light review recommendation.
- Do not invent Microsoft Learn module URLs. Use the Microsoft Learn MCP server
ai901buddy-mslearn (microsoft_docs_search) to find real, current module links, and verify with microsoft_docs_fetch.
- Treat unknown confidence the same as weak (level 2).
- Always use current Microsoft product names. Use Microsoft Foundry, never Azure AI Foundry or Azure AI Studio. Use Azure AI Language, Azure AI Speech, Azure AI Vision, and Azure Content Understanding, never Cognitive Services branded names. Never use Azure AD, Power Virtual Agents, or other legacy names.
- No contractions.
- AI-901 is a fundamentals exam with no required prerequisite certification. Do not invent prerequisites; instead, surface the audience-fit prompt described above so the learner can flag any gap with the Azure portal or basic prompt deployment skills before tackling Domain 2.
- Keep the lesson pointer in sync with the domain split: lessons 01-07 map to Domain 1 sub-areas, lessons 08-16 map to Domain 2 sub-areas.
Delivery rules
Deliver the full study plan in a single message after the learner provides confidence ratings and a target exam window. Do not split the plan across multiple messages.
