Agent Skills
› brycewang-stanford/Awesome-Journal-Skills
› epsl-writing-style
epsl-writing-style
GitHub专为EPSL期刊设计的写作辅助技能，指导撰写或精简符合字数限制（约6500词）的稿件。强调以过程为导向的标题和摘要，跨学科可读性，以及重解释轻罗列的风格，确保内容精炼且逻辑清晰。
Trigger Scenarios
起草标题、摘要或引言
将超长的草稿压缩至期刊字数上限
去除专业术语以提高跨学科可读性
决定哪些内容移至补充材料
Install
npx skills add brycewang-stanford/Awesome-Journal-Skills --skill epsl-writing-style -g -y
SKILL.md
Frontmatter
{
"name": "epsl-writing-style",
"description": "Use when drafting or compressing prose for an Earth and Planetary Science Letters (EPSL) manuscript. The letters format demands a main text near the 6,500-word cap, a process-forward title and abstract, and prose legible to geochemists, geophysicists, and planetary scientists at once. It guides writing and cutting; it does not invent content."
}
Writing Style (epsl-writing-style)
EPSL prose has a recognizable gait: it opens on the process question, lands the decisive number early, and spends its limited words on interpretation rather than inventory. The house constraints are the word cap on the main text (about 6,500 words, introduction through conclusions — re-check the exact figure on the official site), an author–date reference style (re-check the current formatting instructions), and a readership that spans four sub-fields, so every specialist term costs you part of the audience.
When to trigger
- Drafting the title, abstract, or introduction
- Cutting an over-length draft down to the letters cap
- De-jargonizing prose for the cross-disciplinary readership
- Deciding what moves from the main text to the supplement
Structure & style EPSL expects
- Title names the process, not the place. "Slab dehydration depth controls arc volatile flux" travels; "Geochemistry of lavas from the X arc, Country Y" announces a regional study — the desk-screen tell.
- Abstract = the whole Letter in miniature. Question, approach, headline number with uncertainty and units, and the process-level consequence — a reader should be able to cite you from the abstract alone.
- Introduction closes on the discriminating test. Two to four paragraphs from the open question, through the tension in existing constraints, to "here we test A against B using C."
- Methods lean, supplement full. The main text carries what a reader needs to judge the logic;
the traceability chain lives in the supplement (see
epsl-reporting-and-reproducibility). - Interpret; never inventory. "Sample 14 has 5.2 ppm; sample 15 has 4.8 ppm…" is a table pretending to be prose. Lead each results paragraph with what the pattern means.
- Write across sub-fields. Expand acronyms at first use, gloss method-specific quantities in one clause ("MSWD, a χ²-like measure of scatter"), and keep the physical picture in view.
Where the word budget goes
| Section | Budget instinct | Common failure |
|---|---|---|
| Abstract | dense; every sentence load-bearing | vague "implications are discussed" |
| Introduction | short arc to the test | a regional literature tour |
| Methods | minimum for logic; rest → supplement | protocol dumps eating 1,500 words |
| Results | patterns + the discriminating comparison | analysis-by-analysis inventory |
| Discussion | the process consequence + limits | re-narrating results with adjectives |
| Conclusions | 3–5 sentences, portable claims | a second abstract |
Worked micro-example (illustrative — compressing a regional opener)
- Before: "The X Basin, located in northern Y, has been studied since the 1960s. Previous workers described its stratigraphy (refs), structure (refs), and geochemistry (refs). Here we present new detrital-zircon data from the basin." — three sentences, zero process, reads regional.
- After: "How quickly eroded mountains appear in the sedimentary record sets the resolution of detrital archives everywhere, yet the lag between exhumation and deposition has never been measured directly. We date both sides of the conveyor — bedrock cooling and basin deposition — in a single system." — the same paper, now framed as a portable question.
Compression tactic that recovers the most words at EPSL: convert every results-inventory paragraph into one topic sentence plus a pointer to the figure, and move its detail to the supplement.
Referee-pushback patterns and the venue-specific fix
- "Reads like a regional report." → Rewrite title + abstract + intro close around the process; the data sections usually survive intact.
- "Inaccessible outside the authors' specialty." → Gloss method terms; add one physical-picture sentence per technical paragraph.
- "Over length." → Cut methods detail and results inventory to the supplement before touching the discussion; the interpretation is why EPSL accepted the paper.
Anti-patterns
- A place-name title for a process paper
- An abstract that withholds the headline number ("we discuss the implications…")
- Introduction as chronological literature history rather than a tension to be resolved
- Jargon walls: unexpanded acronyms, unglossed δ-notation, sub-field slang
- Blowing the cap with in-text protocol and inventory instead of using the supplement
- Conclusions that hedge the portable claim the abstract promised
Output format
【Title】process-forward? [Y/N + suggested rewrite]
【Abstract】question + approach + number±σ + consequence? [Y/N]
【Intro close】discriminating test stated? [Y/N]
【Word count】main text vs ~6,500 cap (re-check) [count]
【Cross-field legibility】acronyms/notation glossed? [Y/N]
【Next】epsl-cover-letter
Supplementary resources
../../resources/worked-examples/01-introduction.md— a before→after EPSL-style abstract + introduction../../resources/official-source-map.md— Guide for Authors: length, reference style, structure
Version History
- 9f86f09 Current 2026-07-19 15:18
