Vedomosti Editor

Editing skill for business texts based on Leonid Bershidsky's editorial methodology, developed in the tradition of Russian business journalism at Vedomosti. The skill applies structural, logical, and rhythmic frameworks with a focus on fact-checking, clear argumentation, and publishable endings.

This is not a style imitation tool. It encodes an editorial method, not a personal voice, vocabulary, or manner of writing.

What It Does

vedomosti_editor helps an LLM edit or draft business journalism, opinion columns, executive commentary, analytical notes, and corporate-media texts. It is designed for work where the editor must strengthen the argument before polishing the prose.

The skill checks whether a text has:

• one clear thesis;
• a lead that hooks the reader and a second paragraph that sets the contract;
• verifiable facts, figures, and source attribution;
• an argument built from evidence rather than assertion;
• a visible counterargument;
• readable sentence and paragraph rhythm;
• a headline, deck, and ending that do real editorial work.

Method

The skill is based on a reconstructed editorial method from a corpus of Russian-language texts by Leonid Bershidsky from 1999-2014, including business columns, essays, and authorial notes on craft. The methodology separates reusable editorial principles from individual style.

The working model is:

1. Establish the genre, reader, thesis, and target length.
2. Verify facts before restructuring prose.
3. Build the text around one argument.
4. Prove the thesis from the ground up: case, context, objection, conclusion.
5. Adjust rhythm with measurable sentence and paragraph targets.
6. Replace weak headlines, repetitive leads, and summary endings.
7. Return both the edited text and an editor's note explaining the intervention.

Repository Description

Use this as the GitHub repository description:

Editing skill for business texts based on Leonid Bershidsky's methodology, who pioneered business journalism in Russia at Vedomosti newspaper. It applies structural, logical, and rhythmic frameworks with a focus on fact-checking and clear argumentation. The skill strictly encodes the editorial method, not personal style or language.

Suggested shorter alternative:

LLM editing skill for business texts: fact-checking, structure, argument, rhythm, and endings based on Bershidsky's editorial method.

Suggested Topics

llm-skill, editing, editorial-method, business-journalism, fact-checking, argumentation, russian-language, writing-tools, ai-agents

Files

• SKILL.md - the main skill instructions.
• references/methodology.md - the method note distilled from the attached research memo.
• references/editing-checklist.md - operational editorial checklist and scoring rubric.
• references/rhythm-and-syntax.md - measurable rhythm and syntax targets.
• scripts/text_metrics.py - deterministic text rhythm metrics.
• agents/openai.yaml - UI metadata for skill lists.
• examples/prompts.md - example prompts for editing, auditing, and drafting.

How To Use

Install the folder as an LLM skill, then call it with prompts such as:

Edit this business column using the Vedomosti Editor skill. Return the full edited text and an editor's note.

Audit this draft for facts, structure, argument, rhythm, headline, lead, and ending. Do not rewrite yet.

Rewrite this executive comment as a publishable business-media column. Preserve facts, mark any added facts, and explain the edits.

Output Format

The skill returns:

1. Edited text ready for publication.
2. Editor's note:
   • key changes by block: facts, structure, logic, rhythm, headline/lead, ending;
   • before/after score using the 10-point rubric;
   • text_metrics.py results when a local text file is available.

Ethical Boundary

The skill must not imitate Leonid Bershidsky's personal style. It must not reproduce his distinctive wording, voice, or rhetorical habits. It should apply only transferable editorial principles: factual density, source discipline, argument structure, rhythm, and endings.

Method, not mask.