Agent Skills
› brycewang-stanford/Awesome-Journal-Skills
› ngeo-results-framing
ngeo-results-framing
GitHub用于优化地球科学成果叙事，确立单一核心突破，重构摘要与开篇结构。适用于数据已就绪但亮点模糊、叙述像流水账或需强化跨学科意义的场景，不执行分析或绘图。
Trigger Scenarios
有数据模型输出但核心结论模糊
草稿像考察报告而非单一论点
有多个发现未选定 headline
开篇被文献或背景淹没
scope-fit 判定为弱广度框架
Install
npx skills add brycewang-stanford/Awesome-Journal-Skills --skill ngeo-results-framing -g -y
SKILL.md
Frontmatter
{
"name": "ngeo-results-framing",
"description": "Use when a Nature Geoscience result is correct but its headline advance and cross-disciplinary significance are not yet sharp. Structures the one-advance narrative and the \"Here we show\" abstract; does not run analysis or design figures."
}
Nature Geoscience Results Framing (ngeo-results-framing)
When to trigger
- You have the data/model output but the "so what for the Earth system" is fuzzy
- The draft reads like a chronological expedition or model-run report, not a single argument
- You have several findings and have not chosen the headline one
- The opening buries the result under a literature or regional-setting tour
ngeo-scope-fitreturned a Nature Geoscience verdict but flagged the breadth framing as weak
The one-advance discipline
A Nature Geoscience Article makes one headline advance. Everything else is support. Decide the headline result first; demote the rest to corroboration, to the online Methods, or to Supplementary Information.
- Primary advance — the single sentence the title and abstract both encode, phrased as what the Earth system does, not what you measured.
- Supporting results — at most two or three, and only if they reinforce the primary one.
- Everything else — regional context, extended sensitivity tests, secondary proxies → SI (see
ngeo-supplementary).
If you have two co-equal advances, you likely have two papers, or one belongs in a community journal.
The abstract: the "Here we show" convention
Nature abstracts follow a recognizable arc and are capped at 200 words, unreferenced (verify). The house pattern:
- Background / why the question matters — one to three sentences a non-specialist Earth scientist grasps.
- The gap or controversy — what was unknown, unresolved, or assumed.
- "Here we show ..." — the headline advance, stated plainly with the key quantitative result and its uncertainty.
- Mechanism / interpretation — what the result means physically.
- Broad implication — the consequence for the wider Earth system or a neighboring field.
Lead into the advance with the literal "Here we show" (or a close equivalent); Nature editors and readers expect it, and it forces finding-first framing.
The opening (first paragraphs carry the gate)
The main-text opening front-loads significance, not setting. A reliable arc:
- The Earth-system context — one or two sentences on the first-order process at stake.
- The open question / controversy — the specific unknown, framed as a scientific problem, not a regional description.
- What this study shows — the headline advance with the key number.
- Why it matters across the field — the cross-disciplinary consequence.
Do not open with the study region's geography or a paragraph of instrument history. The advance should be legible within the first paragraph.
Framing decision branches
| Symptom | Move |
|---|---|
| Two co-equal headline advances | Split, or demote one to support / SI |
| Result stated only operationally ("we measured X") | Reframe as the Earth-system conclusion it implies |
| Significance asserted, not shown | Tie it to a first-order question or a prior belief it changes |
| Regional setting dominates the opening | Cut to 1–2 sentences; move detail to Methods / SI |
| Abstract lists methods before findings | Reorder: "Here we show ..." first, method as an enabling clause |
| Overreaching climate/impact claim | Calibrate to what the evidence and uncertainty license |
Checklist
- Exactly one primary advance, expressible in one sentence
- Title (≤ ~90 characters) encodes the advance, not just the topic or region
- Abstract uses "Here we show" and leads with the finding + its uncertainty
- Abstract is unreferenced and within the ~200-word limit
- Opening reaches the result within the first paragraph or two
- Significance is tied to a first-order question or belief it changes
- Interpretation (mechanism) is stated, not just the observation
- Supporting results (≤3) genuinely reinforce the primary one
Anti-patterns
- A "kitchen-sink" Article that reports every measurement the campaign produced
- Operational framing ("we observe X in region Y") with no mechanistic interpretation
- Significance claims that overreach beyond the data and its uncertainty
- An abstract that is a methods or fieldwork summary rather than a result statement
- A title naming the region and method instead of the advance
- Saving the punchline for a "conclusions" paragraph the reader may never reach
Output format
【Primary advance】one sentence (Earth-system framing)
【Title】≤ ~90 chars, encodes the advance? yes / fix
【Abstract】"Here we show" + finding + uncertainty, ≤200 words, unreferenced?
【Opening moves】context / gap / advance / broad significance — all present?
【Supporting results】≤3, each justified
【Demoted to Methods / SI】list
【Next】ngeo-methods (grounding) or ngeo-figures (lead figure)
Abstract and title conventions are durable; verify the current word/character limits and the "Here we show" guidance on the official Nature Geoscience author pages (Checked: 2026-07-16).
Version History
- 9f86f09 Current 2026-07-19 17:08
