Skip to content
Get packs

Deep Research course design

Builds courses whose facts are researched, sourced and checked before teaching

Source Download skill
Repository
THU-MAIC/OpenMAIC
Stars
42,460
Licence
MIT
Files
2 files

What it does

When a course depends on recent or externally checkable facts, this skill makes the agent research the topic first: it lists the user's attached materials, runs a small budget of web searches, fetches a few primary sources, and keeps a ledger linking each claim to its source and date. The outline and each page are then written only from those verified claims, with conflicts and gaps stated openly instead of filled with guesses. It is aimed at the course-generation workflow of the OpenMAIC project.

Great for

  • Teachers or course creators who need up-to-date, sourced material on news, policy, markets, products or science
  • Developers building or testing the OpenMAIC course generator who want research-grounded output
  • Students or trainers who want learners to practise telling supported claims from unsupported ones

Skip it if

  • Timeless textbook topics that do not need live sources; the skill itself says not to use it there
  • Anything needing legal, medical or financial advice to be authoritative, since the output is only as good as the few sources fetched within a fixed budget
  • Environments without web search, where it falls back to user materials and general knowledge and cannot verify recent facts

Try asking

  • “Make a 5-page course on how the EU AI Act's latest obligations changed this year, using current official sources”
  • “Build a lesson on recent electric vehicle battery cost trends with sourced figures and the disagreements between reports”
  • “Create a quiz course on the latest version differences of a named software product, checked against its official release notes”
Step-by-step guide: how to use Deep Research course designInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 2 original files, the MIT licence and an attribution note.

    deep-research.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip deep-research.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip deep-research.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Provide your own documents or links up front; the skill treats them as the primary authority and uses web search only to supplement them.
  • Give a specific topic and time frame so the four-facet split and eight-search budget are spent on the questions that matter.
  • Review the ledger or the cited sources in the generated course, since fetched web pages are only as reliable as their origin.

What's inside

  • outline-constraints.json
  • SKILL.md

Redistributed unmodified from THU-MAIC/OpenMAIC under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.