research
Source collection and ideation for lecture materials. After confirming topic/audience/purpose, performs web research (Korean and English by default, adjusted to the engagement's target market and audience when they differ) and writes research_notes.md. Responds to "research topic", "collect sources" . Stage 1 of the lecture workflow.
- License 许可
- AGPL-3.0
- 最近更新
- 2026-11-14
- Stars
- ★ 25
- Forks
- 9
01 前言 Front Matter
- name
- research
- owner
- research
- scope
- co-deck
- status
- active
- outputs
- [ "research-notes" ]
- version
- 1.2.1
- relates_to
- [ { "type": "follows", "skill": "storyline" } ]
- description
- Source collection and ideation for lecture materials. After confirming topic/audience/purpose, performs web research (Korean and English by default, adjusted to the engagement's target market and audience when they differ) and writes research_notes.md. Responds to "research topic", "collect sources" . Stage 1 of the lecture workflow.
- last_reviewed
- 1787443200000
- prerequisites
- none
03 SKILL.md 正文
Context
Researches the lecture topic via web search (Korean + English) and organizes findings into research_notes.md — the primary input for the Storyline Agent. Invoked at Stage 1, after the lecture topic and audience are confirmed.
When to Use
- PM Agent dispatches at Stage 1
- User says "research X"
- Before writing storyline when
research_notes.mddoes not yet exist
Execution Steps
Step 1: Confirm Lecture Information
Confirm the following with the user before starting (skip anything already mentioned):
- Topic — what the lecture is about
- Audience — experts/general public, industry, seniority level
- Purpose — educate, persuade, introduce, inspire, etc.
- Length — slide count or presentation time
- Constraints — must-cover topics, must-avoid topics
Step 2: Research
Use WebSearch and web_fetch to collect:
- Core concepts and definitions of the topic
- Latest trends, market dynamics, statistics
- Concrete case studies
- Expert quotes or reports worth citing
- Real-world examples the audience can relate to
Korean and English by default; adjust to the engagement's target market and audience when they differ. Always record source URLs.
Step 3: Write research_notes.md
Organize collected content into the structure below and save to the workspace.
# Research Notes: [Topic]
## Lecture Overview
- Topic:
- Audience:
- Purpose:
- Length (slide count):
## Core Messages (3-5)
Key points the audience must take away.
1.
2.
3.
## Key Concepts & Content
### [Concept/Section 1]
[Summary]
### [Concept/Section 2]
[Summary]
## Data / Statistics
- [Figure] — Source: [URL]
- [Figure] — Source: [URL]
## Case Studies
### [Case 1]
[Description]
## References
- [Title](URL)
- [Title](URL)
Korean example (when the lecture targets a Korean audience, write actual content in Korean):
# Research Notes: Securities Industry in the AX Era ## Lecture Overview - Topic: Securities Industry Changes in the AX (AI Transformation) Era - Audience: Securities company executives and practitioners - Purpose: Providing insights and strategic directions - Length (Slide count): 60
Output Format
presentations/<project>/research_notes.md with sections: Lecture Overview, Core Messages, Key Concepts & Content, Data / Statistics, Case Studies, References. See Step 3 template above for full structure.
Related Skills
storyline— consumesresearch_notes.mdoutput from this skill
04 同仓库更多 Skills
i18n-layout
Text layout and encoding guidance: character encoding (UTF-8, legacy Korean code pages, BOM hazards), line endings, RTL/bidi handling, and script-specific font selection (Hangul focus). Use when: encoding is corrupted or mojibake appears, an RTL locale must render correctly, fonts must be chosen for a script, or CRLF/BOM issues surface.
i18n-layout
Text layout and encoding guidance: character encoding (UTF-8, legacy Korean code pages, BOM hazards), line endings, RTL/bidi handling, and script-specific font selection (Hangul focus). Use when: encoding is corrupted or mojibake appears, an RTL locale must render correctly, fonts must be chosen for a script, or CRLF/BOM issues surface.
i18n-layout
Text layout and encoding guidance: character encoding (UTF-8, legacy Korean code pages, BOM hazards), line endings, RTL/bidi handling, and script-specific font selection (Hangul focus). Use when: encoding is corrupted or mojibake appears, an RTL locale must render correctly, fonts must be chosen for a script, or CRLF/BOM issues surface.
i18n-layout
Text layout and encoding guidance: character encoding (UTF-8, legacy Korean code pages, BOM hazards), line endings, RTL/bidi handling, and script-specific font selection (Hangul focus). Use when: encoding is corrupted or mojibake appears, an RTL locale must render correctly, fonts must be chosen for a script, or CRLF/BOM issues surface.
i18n-layout
Text layout and encoding guidance: character encoding (UTF-8, legacy Korean code pages, BOM hazards), line endings, RTL/bidi handling, and script-specific font selection (Hangul focus). Use when: encoding is corrupted or mojibake appears, an RTL locale must render correctly, fonts must be chosen for a script, or CRLF/BOM issues surface.
i18n-layout
Text layout and encoding guidance: character encoding (UTF-8, legacy Korean code pages, BOM hazards), line endings, RTL/bidi handling, and script-specific font selection (Hangul focus). Use when: encoding is corrupted or mojibake appears, an RTL locale must render correctly, fonts must be chosen for a script, or CRLF/BOM issues surface.