What problem does it solve? Writing a benchmark or evaluation paper requires a fundamentally different structure than a technical paper, and researchers often miss critical elements reviewers expect, such as a construction pipeline figure, a benchmark comparison table, or bolded Finding summaries. This Skill audits a benchmark idea against a five-pillar framework and produces the complete logic chain and section skeleton before writing begins. ## Core Features & Use Cases - Five-Pillar Completeness Audit: Checks whether the Research Gap, Construction Pipeline, Evaluation Framework, Empirical Findings, and optional Companion Method are each articulated, with improvement suggestions for gaps. - Introduction Six-Part Logic Chain: Generates the Background plus Running Example, Existing-Benchmark Limitations, Research Questions, Design Considerations, Proposal, and Contributions structure specific to benchmark papers. - Section Skeleton and Pre-Submission Checklist: Produces a Section 2-7 outline with figure and table placement, plus a four-category reviewer checklist with Critical, Major, and Minor severity classification. - Use Case: A researcher planning a Text-to-SQL benchmark submits their idea, receives a completeness table showing their evaluation framework lacks a fine-grained taxonomy, and gets a full section skeleton with the pipeline figure and comparison table planned before data construction starts. ## Quick Start Use the phd-benchmark-paper-template skill to audit my benchmark paper idea on ambiguity in code generation and produce the five-pillar completeness table, Introduction logic chain, and section skeleton.
