bama
BAMA: 모든 Mac 앱을 만들 수 있습니다
우리 세계 데이터를 위한 에이전트 스킬
Agent Skills for Our World in Data
스킬Shell · PythonGitHub @owid직접 설치운영자 미확인
써 보셨어요? 한 번만 눌러 주세요. 로그인은 필요 없어요.
아직 감지된 업데이트가 없습니다. GitHub 릴리스가 나오면 여기에 보입니다.
개인 계정의 공개 저장소 소유자는 GitHub로 로그인해 운영자를 확인하고 소개를 직접 관리할 수 있습니다. 관리자 승인은 필요하지 않습니다.
로그인한 뒤 확인 파일을 만들어 이 저장소의 기본 브랜치에 30분 안에 커밋해 주세요. 다른 브랜치나 포크의 파일은 인정하지 않습니다. 파일을 올린 뒤 아래에서 확인하면 관리자 승인 없이 관리할 수 있습니다. 자격을 유지하려면 파일을 남겨 두세요.
로그인이 필요하면 로그인 화면으로 이동합니다. 로그인 후 이곳으로 돌아와 원하는 확인 버튼을 다시 눌러 주세요.
설치형 프로젝트는 배포 도메인 없이도 확인할 수 있습니다. 개인 저장소는 GitHub 로그인으로, 조직 저장소는 위의 저장소 확인 파일로 직접 확인해 주세요.
README 전체를 불러오는 중입니다. 아래는 저장된 발췌입니다.
AI skill for Our World in Data
An agent skill for working with Our World in Data. It teaches your AI assistant to find our charts and articles, pull the numbers behind them, put a chart into a page or a slide, and credit whoever produced the data. It works with the Claude and ChatGPT apps, Claude Code, Codex, Gemini CLI, Cursor, GitHub Copilot and others.README 전문 보기 ↗An agent skill for working with Our World in Data. It teaches your AI assistant to find our charts and articles, pull the numbers behind them, put a chart into a page or a slide, and credit whoever produced the data. It works with the Claude and ChatGPT apps, Claude Code, Codex, Gemini CLI, Cursor, GitHub Copilot and others.
Our World in Data publishes thousands of charts and hundreds of articles on global problems: poverty, health, energy, climate, education, and more. This skill teaches assistants how to use that content properly: how to find the right chart, how to get the numbers behind it, what the data does and does not mean, and how to credit the people who collected it.
Once installed, you can ask your agent things like:
Most of the time you don't have to ask for it by name — your assistant reaches for it on its own when your question is about our charts or data.
The skill is a short set of notes your assistant reads when it needs them: how to
search our site, how to fetch the data and sources behind a chart, and how to put
a chart into a page or a slide. You can read them in
skills/owid/ if you're curious.
It runs no code of its own and needs no API key or account. Your assistant reads public pages and data files on ourworldindata.org, and nothing else.
Our World in Data's own work is free to reuse under the Creative Commons BY licence. Most of the data, though, comes from other producers and keeps whatever licence they set, so check before you republish — and a few charts cannot be downloaded at all. The skill instructs agents to name the original producer, so please keep the citations when you publish results.
Every request the skill makes identifies itself as coming from this skill, which is how we can see it being used and keep supporting it. That is why links it gives you end in utm_source=owid-skills; the FAQ explains.
The quickest way is to ask your AI assistant:
Install the Our World in Data skill hosted at https://github.com/owid/skills/
Claude Code and Codex can do it themselves; in the Claude and ChatGPT apps, your assistant can walk you through the current menus.
What each app needs:
| Your app | How |
|---|---|
| Claude app (paid plan) | In the plugin settings, add a marketplace from owid/skills, then install Our World in Data |
| ChatGPT app | In the plugin settings, add a marketplace from owid/skills, then install Our World in Data |
| Claude Code | /plugin marketplace add owid/skills, then /plugin install owid@owid-skills, then turn on auto-update for owid-skills under Marketplaces in /plugin |
| Codex | codex plugin marketplace add owid/skills, then codex plugin add owid@owid-skills |
| Anything else | npx skills add owid/skills, or copy the skills/owid/ folder into your agent's skills directory |
None of them needs an Our World in Data account or a GitHub account. Start a new chat afterwards: plugins only apply to conversations that begin after installing.
Check it worked. In a new chat, ask:
Use the Our World in Data skill to fetch the data behind this chart: https://ourworldindata.org/grapher/child-mortality — the complete time series for Uganda, and the source the chart relies on.
A working install returns the whole series for Uganda (about seventy yearly figures, starting in the 1950s) and names Gapminder and the UN Inter-agency Group for Child Mortality Estimation as the source, not "Our World in Data". An assistant without the skill will say so; one that has it but didn't fetch anything will summarise or invent the numbers. If that happens, see the FAQ.
Updating. There are no version numbers: we change the skill in place. Update it
where you installed it: reinstall it in the apps, run claude plugin update owid@owid-skills in Claude Code (or turn on auto-update for the owid-skills
marketplace in /plugin), codex plugin marketplace upgrade owid-skills in Codex,
or npx skills update for the skills CLI.
Creative Commons BY 4.0 — the same terms as the rest of Our World in Data's own work. Use it anywhere, change it, ship it inside your own project; just credit Our World in Data.