tutor-research

Tutor Research

Overview

Receives a learning question, investigates multiple paths in paralleland synthesizes reliable, up-to-date learning evidence. It runs two axes simultaneously: ① context7 MCP — official documentation by version for libraries, SDKs, frameworks, and CLIs, ② WebSearch — the latest changes, tutorials, concept comparisons, and real-world cases. It cross-verifies sources to block hallucinations, and learning-materialproduces a structured synthesis that can be used right away.

Trigger Keywords

investigate, research, latest information, official documentation, context7, parallel investigation, accurate details, tell me, compare, difference, how it works

Workflow

Step 1: Analyze the question → derive research axes

Break it down into sub-axes to answer the question.

• Official documentation axis (context7 target): Does a specific library, SDK, tool, or API appear? (e.g., React 19, Next.js, claude code CLI, FastAPI) → extract resolve-library-id targets
• Latest information axis (WebSearch target): concepts, trends, comparisons, "when to use what," recent changes → 2-4 search queries
• Codebase axis (optional, Explore): when including code inside the user's project as a learning target

Step 2: Execute parallel research

[HARD] Independent investigations are run in parallel in one turn (serial prohibited). The orchestrator calls the following simultaneously.

• context7: resolve-library-id → get-library-docs (relevant library/topic)
• WebSearch: 2-4 derived queries
• (when the topic is broad) Agent(Explore) or Agent(general-purpose) researchers for each sub-axis in parallel fan-out — each agent takes one axis and returns investigation results

For broad, multi-faceted topics (where source cross-verification is key), you can propose the bundled /deep-research <질문> workflow. However, first inform the user that the token cost is greater than a single search.

[한 턴 안에서 병렬 호출]
  ├─ context7: get-library-docs(<라이브러리>, topic=<핵심 토픽>)
  ├─ WebSearch: "<주제> 2026 최신"
  ├─ WebSearch: "<개념 A> vs <개념 B> 차이 언제 사용"
  └─ (선택) Agent(Explore): 사용자 프로젝트 내 관련 코드 패턴

GLM backend caution: moai glm or moai cg's GLM pane, WebSearch and WebFetch are routed to z.ai MCP tools. .claude/rules/moai/core/glm-web-tooling.mdfollow the routing table in

Step 3: Cross-verify + synthesize

• Source verification: For WebSearch result URLs, cite only those whose existence and content have been confirmed via WebFetch. Do not generate URLs that are not in the search results (anti-hallucination policy).
• Conflict handling: If official documentation and web sources differ, prioritize official documentation (context7), and mark web sources as supplementary/contextual. For conflicts caused by version differences, specify the version.
• Level customization: Adjust terminology difficulty to match the learner level of the learning-project (for beginners, start from prerequisite concepts).

Step 4: Output the comprehensive document (learning-material input specification)

learning-materialCreate structured Markdown that can be consumed right away.

## 리서치 종합: <주제>
- 학습자 수준: <입문/초급/중급/고급>

### 핵심 개념 (요약)
- 개념 1 — 한 줄 정의 + 근거 출처
- ...

### 도식 후보 (mermaid 소스 초안)
- 개념 관계도 / 플로우 / 시퀀스 중 적합한 것 1-2개 초안

### 예제·코드 (출처·버전 표기)
- 공식 문서 기준 최소 예제

### 자주 헷갈리는 점 / 비교
- A vs B, 언제 무엇을 쓰나

### 출처
- 공식 문서: <context7 라이브러리·버전>
- 웹: <검증된 URL 목록>

Deliverables

• Structured research comprehensive document (Markdown, with sources)
• Mermaid diagram source draft (passed to learning-material)
• List of verified sources (official documentation + web URLs)

Usage examples

Example 1: "Research the latest information on how Claude Code subagents work" → context7 (Claude Code docs) + WebSearch (latest changes) in parallel → comprehensive document + sequence diagram draft

Example 2: "Prepare an explanation of the React 19 use hook based on official documentation" → context7 (React 19) as the main source + WebSearch (practical examples) → comprehensive document with examples and comparisons

Example 3: "Research the accurate rules and example sentences for the English past perfect subjunctive" → WebSearch (grammar rules and example sentences) in parallel → comprehensive document with a rules table + example sentences (library-independent → skip context7)

Precautions

• Do not make assertions without sources: Mark unverified content as "needs verification".
• When using WebSearch, a Sources section is mandatory — Include only verified URLs as Markdown links.
• Fallback when context7 is unavailable: If the MCP does not respond, directly retrieve the official documentation site with WebFetch.
• Do not ask the user directly: If additional information is needed, leave it as a "needs verification" item in the comprehensive document, and the orchestrator handles decisions via AskUserQuestion.

Related skills

• moai-tutor:learning-project: The milestone's key questions become the input for this skill
• moai-tutor:learning-material: Convert this skill's comprehensive document into diagrams and HTML learning materials

When not to use this skill

• Literature review and citation management for academic papers → moai-education:research-assistant
• Market and business research → moai-business:market-analyst
• Deep multi-source cross-verification reports (single question, non-learning) → bundle /deep-research workflow