Browse reusable Agent Skills, each with a clear purpose and practical guidance.
Use when responding to 《世界经济》 (The Journal of World Economy) referee reports (R&R) — structuring a point-by-point response letter that addresses the journal's characteristic challenges (open-economy fit, shift-share / gravity inference, transmission channel, GVC measurement). Use after the manuscript has actually been revised.
Use when a 《世界经济》 (The Journal of World Economy) manuscript stops at "it promoted exports" without showing how an external shock propagates — making the trade / investment / financial / expectation channel explicit and testable, not a hand-wavy mechanism paragraph.
Use when drafting KDD rebuttals on OpenReview, where authors answer each review under a no-hyperlink rule, cannot upload revisions, and write for the area chair who recommends Accept, Reject, or Resubmit. Covers evidence-anchoring without links, per-reviewer triage, and turning a likely Resubmit into next-cycle material at ACM SIGKDD.
Use when revising a KDD paper into the venue's register, where the first page names a data regime, a mechanism, and scale evidence, efficiency adjectives trace to design decisions, and two-column sigconf pages punish sprawl. Covers Research vs ADS voice, dataset-size discipline, assertive contribution bullets, and 8-page compression.
Use to write The Lancet's structured abstract — ≤300 words under the exact headings Background, Methods, Findings, Interpretation, Funding — leading with the clinical question, reporting the primary outcome with effect size and 95% CI, citing the registration number, and naming the funder.
Use after Lancet reviews arrive (including the statistical reviewer and the editor's letter) to triage the decision, answer statistical and clinical comments, and draft a point-by-point response that quotes the reviewer, gives evidence, quotes the revised manuscript text, and keeps causal language cautious. Do not run before the main text is actually revised.
Use to build The Lancet's signature Research in context panel — the three mandatory headed parts (Evidence before this study, Added value of this study, Implications of all the available evidence), grounded in a systematic literature search with stated databases, terms, and dates. Mandatory for original research Articles.
Use to enforce The Lancet's clinical-statistics reporting — confidence intervals over bare P values, a pre-specified primary analysis, intention-to-treat with per-protocol sensitivity, multiplicity control for secondary endpoints, cautious pre-specified subgroup analyses with interaction tests, missing-data handling, and absolute plus relative effects.
Use to structure and hold the length of a Lancet Article main text — Introduction, Methods, Results, Discussion (~3000–3500 words), the Research in context panel, a limited reference list (~30), and a cautious, globally minded Discussion that does not overstate.
Use when positioning a Language (LSA) manuscript within the literature so it reads as in dialogue with current linguistics across frameworks. Language reviewers are drawn from across subfields; the positioning must engage the strongest rival accounts fairly, not build a subfield-only wall. Maps the conversation; it does not write the literature review.