Bulk Ozon and Wildberries title optimizer skill
russian-title-optimizer
A skill reads a column of titles from Excel and generates a standard and an ad-style Russian title for each, for Ozon and Wildberries
- 3 ★ on the repository
- Last commit
- MIT
Minimal access
Why this level
- Only works with an Excel file on disk, does not call any marketplace API
Install
- Works in
- Services
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/russian-title-optimizerDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewMinimal accessThe tool runs nothing and makes no network calls.
- Only works with an Excel file on disk, does not call any marketplace API
This is third-party code. Review the repository files before installing.
What it does
The script reads source product titles from column A of an Excel file's first sheet and appends two columns per row: a standard title and an ad-style title in Russian. Along the way it strips repeated words, mixed languages, leftover translation fragments and unmotivated marketing words, while keeping product type, material, size, quantity, model, compatibility and purpose. Identical source titles are processed once thanks to a cache, and the result is reused for duplicates. Blank rows and rows with a single digit are skipped unchanged.
Who it is for. For Ozon and Wildberries sellers listing hundreds of products who want to bulk-clean titles of clutter.
Good fit when
- You need to bulk-fix titles after machine translation or scraping from 1688
- You want two title variants at once: standard and ad-style, matching the platform's limits
Not a fit when
- You need keyword research by frequency, not wording cleanup: the skill does not analyze demand
- Titles are already hand-written by a native speaker and need no cleanup
Example request
Process this file of product titles and generate standard and ad-style variants for OzonLimitations
The skill only rewrites text; it does not verify the title against the product's real specs and does not call any marketplace API. If a source title cannot be meaningfully recovered, the row is left as-is with no new titles and no explanation. Documentation is in Chinese, with an openpyxl dependency.
How to disable. Remove the skill from your agent's skills directory.
FAQ
What happens with duplicate titles?
The skill processes a unique title once and reuses the result for every duplicate, speeding up large files.
What if a title cannot be parsed?
The row is kept as-is, with the standard and ad-style title columns left blank.
Discussion
No questions or reviews yet.
Related
All Ozon toolsBulk Ozon and Wildberries description compliance checker
russian-description-compliance
A skill rewrites product descriptions into clean Russian HTML, removing risky marketing claims and translating non-Russian text
Ozon keyword research and listing optimizer
Ozon 搜索词调研与商品卡优化工具
A skill turns a search-query export from the Ozon seller dashboard into an Excel keyword report and a Word title-optimization report
A pair of Codex skills turns product photos and confirmed data into a title, a 3000-4000 character description and tags for Ozon, checked against Wordstat
Ozon Card Optimizer: listing pipeline
ozon-card-optimizer
A pipeline skill for Ozon listings: from verified product facts through Russian SEO copy and interest tags to a six-image set
Russian listing copy for WB and Ozon
ru-listing-writer
An Agent Skill for Claude Code and Codex that writes a Russian listing title and description from product photos and keyword screenshots, per WB and OZON rules
Ozon auto-listing skill
auto-listing-skill
A Codex skill locates the shop's working folder, fills price and category workbooks, builds an image zip, and logs the Ozon auto-listing task
README in short
The Chinese README describes functions line by line: where data is read from, which three new columns are created, which text flaws are cleaned and what is preserved. A separate section covers processing optimizations: caching by source title and rule version, skipping blank and single-digit rows, and reprocessing only failed rows.