Browse reusable Agent Skills, each with a clear purpose and practical guidance.
在为《计算机研究与发展》(Journal of Computer Research and Development, JCRD) 稿件构建可复现性与实验可重复证据时调用。覆盖环境与依赖版本固定、随机种子与非确定性控制、数据集版本与抽取日期钉死、大模型/API 输出缓存、硬件与度量口径记录、双盲下的匿名可运行材料、诚实的可用性说明，以及让外审专家在不联系作者情况下重跑主结果的组织方式。适用于把一份 JCRD 中文稿件的实验做到可被第三方在受控条件下重复、并经得起外审可复现质疑的场景。
在需要理解或建模《计算机研究与发展》(Journal of Computer Research and Development, JCRD) 审稿流程、预估周期与决定作者应对策略时调用。覆盖编辑初审→专家外审→复审→主编终审的多轮结构、双盲评审机制（外审只见内容不见身份）、录用/退修/退稿的决定类型、多轮修回的节奏、专题稿件由客座编辑参与的流程、以及作者在各环节的杠杆点与常见状态解读。适用于收到某一审稿状态后校准预期、或在投稿前了解本刊评审如何运作以便针对性准备的场景。
Use when estimation, heterogeneity, attrition, measurement, or inference choices need to meet Journal of Development Economics (JDE) empirical norms — clustered field data, survey measurement error, and treatment-effect heterogeneity in low- and middle-income settings. Covers the analysis itself, not the identifying design.
Use when the causal identification strategy is the bottleneck for a Journal of Development Economics (JDE) manuscript — RCT/field experiment, DID, IV, RDD in low- and middle-income settings. Stress-tests the design against development-economics empirical norms before tables are drafted.
Use when choosing a submission route or understanding how peer review works at the Journal of Development Economics (JDE) — the full-length, AER-Insights-style short-paper, and pre-results review (Registered Reports) tracks, single-anonymized review, timelines, and the submission cap. Explains the pipeline; it is not the final preflight.
Use when the design that isolates a behavioral mechanism is the bottleneck for a Journal of Economic Behavior & Organization (JEBO) manuscript — lab/field experiment, observational causal design, or simulation. Stress-tests experimental and observational identification to JEBO's behavioral-credibility bar; it does not write prose or build the deposit.
Use when a Journal of Economic Behavior & Organization (JEBO) result may be fragile to demand effects, multiple comparisons, specification, or tuning. Organizes robustness by the behavioral threat each check addresses — for experiments, observational designs, and simulations; it does not redesign the identification or write the prose.
Use when deciding whether a behavioral, experimental, or organizational question fits the Journal of Economic Behavior & Organization (JEBO) rather than a sibling outlet, and how to frame it. Sharpens scope and the behavioral hook; it does not invent evidence, editors, or citations.
Use when analyzing and reporting results for a Journal of Educational Psychology manuscript. JEP expects analyses that respect nesting (multilevel/SEM/growth models), report educationally meaningful effect sizes with confidence intervals, test mechanisms (mediation/moderation), and follow JARS with full disclosure. Guides analysis norms; it does not fabricate results.
Use when meeting the Journal of Educational Psychology's open-science requirements — the Transparency and Openness subsection stating data/materials/analysis-code availability with persistent identifiers, JARS-compliant reporting, preregistration disclosure (encouraged, not mandatory), and Open Science Badges. Prepares compliance for masked review; it does not waive requirements.