Who should read this — Authors defining Research and Seed contracts. Use with Output JSON Scheme for pipeline output structure.
Category indicates the nature of the content (What it is). It is defined based on the essence of the information covered, not the source.
| Category | Meaning Definition | Content Included |
|---|
| news | Timely official/unofficial announcements | Policy announcements, notices, press releases |
| policy | Systems, laws, regulations themselves | Legal amendments, administrative guidelines |
| guide | Content for Explanation/Guidance | How-to, Procedure Guides |
| faq | Q&A Structured Information | FAQ, Help, Support |
| data | Numerical/Statistical/Raw Data | Statistics Korea, Reports |
| research | Academic/Research | Papers, White Papers |
| announcement | Simple Notices | Events, Schedules |
| directory | List-type Information | Agency Lists, Contacts |
| opinion | Commentary·Columns | Contributions, Analysis |
| alert | Emergency/Caution Information | Disaster, Warning |
At the Seed stage, determine nature as one or more of the above values.
| Concept | Where to Place |
|---|
| Government | source.type = "government" |
| Organization | source.type = "organization" |
| NGO / Intl Org | source.type = "ngo" |
| Company | source.type = "company" |
| Medium | Definition |
|---|
| web | General web page |
| social | SNS post |
| video | Video content |
| audio | Podcast |
| document | PDF, DOC |
| Engine | Role |
|---|
| Research Engine | ”Where to look” (URL discovery) |
| Seed Engine | ”How to fetch” (contract creation) |
| Scraper | ”Actual collection” (execution) |