Edit mathematically, empirically, or technically dense papers for readers outside the author’s specialty, including empirical legal scholarship, law-and-economics models, and legal-technology research. Clarifies terminology, antecedents, study design, and claim–statistic relationships while checking equations, numbers, quotations, and citations against the source. Use to clarify exposition or de-jargonize a technical draft; not to rewrite briefs or doctrinal arguments. Produces an edited draft and audit memo. Full integrity checks and original-format rendering require suitable extraction, code execution, and document tools; otherwise provides proposed edits with verification limits disclosed.
Reader-first technical edit
Overview
A technically sound paper can still leave readers guessing what its symbols mean, which sample a result describes, or why a section matters. Reader-first technical edit repairs those problems while keeping the author's technical content under an explicit preservation check.
The skill serves researchers, law professors, students, and lawyers working with empirical legal studies, law-and-economics models, or legal-technology evaluations. It also retains its general use for mathematical, empirical, and systems papers. It edits technical exposition in mixed papers; briefs and doctrinal arguments require a different editing method.
The workflow names the intended reader, inventories fixed content, and runs nine editing passes covering object names, antecedents, definitions, section purpose, compressed reasoning, statistical claims, jargon, actors, and drafting residue. Formal papers receive notation checks; empirical papers receive a study-design map and clearer denominators and selection rules. Legal quotations, citations, defined terms, and jurisdictional qualifications remain protected.
The result is an edited draft plus an audit memo distinguishing changes, unresolved source problems, and completed verification. When the host supports extraction and code execution, the skill compares equations, numbers, quotations, and citations before and after editing. An unchanged inventory is evidence of preservation, not proof that the research or every revised sentence is correct.
Example requests
- “Make this empirical legal paper readable for a law professor who does not specialize in statistics. Preserve the estimates and citations.”
- “Clarify this law-and-economics model for an intelligent lawyer. Explain the notation before using it.”
- “Edit the methods and results of this legal-AI benchmark. Make every headline number’s denominator and selection rule clear.”
What to supply
Provide the draft, the intended audience, and any house style rules. Editable source is preferable; scanned pages may need additional inspection. Identify any permitted renames or substantive changes.
What you receive
An edited document in the original format when supported, a rendered version when available, and a short audit memo covering the audience, defects, examples, changes, unresolved issues, and verification limits. Without suitable tools, the skill supplies proposed edits rather than claiming a fully verified rewrite.
Requirements and limits
No connector or companion skill is required. Complete mechanical verification requires extraction and code execution; original-format delivery requires compatible editing and rendering tools. The skill does not validate the research, check whether authorities remain good law, or provide legal advice. The editing method travels across jurisdictions; substantive law does not.
Files and license
SKILL.md contains the instructions. NOTICE records provenance. LICENSE supplies the Apache-2.0 terms. Author: Seth J. Chandler.