What is the Wildberries Seller API skill?

It packages the Wildberries Seller API's OpenAPI/Swagger documentation as a local, agent-searchable knowledge source, allowing Claude Code, Codex, or other agents to answer questions about WB seller endpoints without network access or loading the full spec into context.

What lookup modes does the wb_api.py helper provide?

The helper offers several modes: `map` for an overview, `search` for keyword search, `tag` for tag-filtered browsing, `detail` for full request/response schema expansion, `protocol` for protocol-level details, and `generate` to produce markdown reference docs.

Which agent platforms does this skill work with?

The skill is agent-agnostic and installs into Codex (user or repo-scoped skill directories) and Claude Code (personal or project skill directories) following the community agentskills.io package format.

When should I use this skill?

Use it whenever a task touches Wildberries seller integrations such as product cards, prices, stocks, orders, supplies, feedbacks, analytics, reports, promotion, tariffs, or finances.

Skill

Integrate with Wildberries Seller API

Name: Wildberries Seller API
Availability: OnlineOnly
Author: NeuralDeep

Agent Skill that gives Claude Code and Codex offline lookup access to the Wildberries Seller API's Swagger specs - endpoints, schemas, and validation.

Get skill

Works with wildberries

NeuralDeep

Own this? Claim it

Maintainer of this project? Claim this page to edit the listing.

Spark score

out of 100

Updated 2 months ago

Version 1.0.0

Models

claudeuniversal

Add to Favorites

Why it matters

Seamlessly connect to the Wildberries Seller API to manage your e-commerce operations. Access comprehensive data across content, pricing, orders, analytics, and more.

Outcomes

What it gets done

Access all Wildberries Seller API sections including content, prices, and orders.

Retrieve analytics and financial data for business insights.

Manage promotions and buyer chat interactions.

Automate data extraction for reporting and analysis.

Install

Add it to your toolbox

Run in your project directory:

curl -fsSL https://spark.entire.vc/get/wb-seller-api-skill | bash

Overview

Wildberries Seller API

Gives coding agents like Claude Code and Codex offline, searchable access to the Wildberries Seller API through local Swagger files. A helper script looks up endpoints, tags, and schemas, and validates the specs, without making network calls. Use it whenever a task involves Wildberries seller integrations - product cards, prices, stocks, orders, supplies, feedbacks, analytics, reports, promotion, tariffs, or finances. Russian Swagger files are canonical for schemas and protocol; the English mirror is for search and labels only.

What it does

This skill packages the Wildberries Seller API's OpenAPI/Swagger documentation as a local, agent-searchable knowledge source, so Claude Code, Codex, or any agent that loads SKILL.md can answer questions about WB seller endpoints without needing network access or loading the full spec into context. The canonical source of truth is a set of Russian Swagger YAML files (swagger/*.yaml); a parallel English mirror (swagger/en/*.yaml) is layered on top purely for search and human-readable labels.

When to use - and when NOT to

Use it whenever a task touches Wildberries seller integrations: product cards, prices, stocks, orders, supplies, feedbacks, analytics, reports, promotion, tariffs, or finances. Agents can trigger it automatically on relevant prompts (e.g. "How do I fetch Wildberries product cards?" or "Найди endpoint для рейтинга продавца") or be invoked directly, for example via /wildberries-api in agents that expose skills as commands. It is not a live API client - the bundled helper never makes network calls, so it cannot execute requests against Wildberries; for exact request parameters, auth schemes, and hosts, treat the Russian Swagger as authoritative and the English overlay as a convenience layer only, since it can drift structurally from the canonical spec.

Inputs and outputs

After installing the Python dependency (pip install -r requirements.txt), the bundled scripts/wb_api.py helper offers several lookup modes run from the skill directory: map for an overview, search "seller rating" --limit 5 --lang en (or a Russian query without --lang) for keyword search, tag "Product Cards" --lang en for tag-filtered browsing, detail POST /content/v2/get/cards/list --schemas for full request/response schema expansion on one endpoint, protocol for protocol-level details, and generate --output generated --lang en to produce markdown reference docs on demand. Output is always compact, locally derived text - never a live network response.

Integrations

The skill is agent-agnostic by design, following the community agentskills.io package format: it installs the same way into Codex's user (~/.agents/skills/) or repo-scoped (.agents/skills/) skill directories, and into Claude Code's personal (~/.claude/skills/) or project (.claude/skills/) skill directories. A validate command checks the health of the underlying specs - that both language variants parse, that there are no duplicate method/path pairs or unresolved $ref values, that operations have titles and tags, and that Russian/English coverage and generated-docs freshness stay in sync.

Who it's for

Developers building or maintaining integrations with the Wildberries marketplace who want their coding agent to answer API questions and generate request code accurately from the real spec, instead of guessing at endpoint shapes from training data.

Source README

Wildberries Seller API - Agent Skill

An Agent Skills package for working with the Wildberries Seller API in compatible coding agents such as Codex, Claude Code, and other tools that load SKILL.md.

The repository keeps local Swagger files as the single source of truth for API documentation. Russian specs in swagger/*.yaml are canonical for schemas and protocol details; English specs in swagger/en/*.yaml are used as a validated search and text-label overlay.

Installation

Clone this repository into the skills directory used by your agent.

Codex user skill:

git clone https://github.com/miilv/wb-seller-api-skill.git ~/.agents/skills/wildberries-api

Codex repository-scoped skill:

git clone https://github.com/miilv/wb-seller-api-skill.git .agents/skills/wildberries-api

Claude Code personal skill:

git clone https://github.com/miilv/wb-seller-api-skill.git ~/.claude/skills/wildberries-api

Claude Code project skill:

git clone https://github.com/miilv/wb-seller-api-skill.git .claude/skills/wildberries-api

Restart or refresh your agent if it does not detect the new skill automatically.

Usage

Agents can activate the skill automatically when you mention Wildberries API, WB seller integrations, marketplace endpoints, product cards, prices, stocks, orders, supplies, feedback, analytics, reports, promotions, tariffs, or finances. You can also invoke it directly in agents that expose skills as commands, for example /wildberries-api.

Example prompts:

How do I fetch Wildberries product cards?
Find the endpoint for seller rating.
Generate a Python client call for FBS order metadata.
Which WB API host should I use for analytics reports?
Show request and response schemas for POST /content/v2/get/cards/list.

Included Files

File	Description
`SKILL.md`	Skill instructions and progressive-disclosure workflow
`swagger/*.yaml`	Canonical Russian OpenAPI specs for Wildberries Seller API
`swagger/en/*.yaml`	English Swagger mirror used for search and text labels
`scripts/wb_api.py`	Local helper for Swagger-derived maps, endpoint search, detail lookup, schema expansion, generated docs, and validation
`requirements.txt`	Python dependency list for the helper

Lookup Helper

Install the helper dependency if needed:

python3 -m pip install -r requirements.txt

Run compact Swagger-derived lookups from the skill directory:

python3 scripts/wb_api.py validate
python3 scripts/wb_api.py map
python3 scripts/wb_api.py search "seller rating" --limit 5 --lang en
python3 scripts/wb_api.py search "рейтинг продавца" --limit 5
python3 scripts/wb_api.py tag "Product Cards" --lang en
python3 scripts/wb_api.py detail POST /content/v2/get/cards/list --schemas
python3 scripts/wb_api.py detail POST /content/v2/get/cards/list --schemas --lang en
python3 scripts/wb_api.py protocol
python3 scripts/wb_api.py generate --output generated --lang en

The helper does not make network calls and is not a Wildberries API client. It reads local Swagger files and prints compact output so agents do not need to load full API specs into context.

--lang en changes human-readable labels and descriptions where an English mirror exists. Exact parameters, request bodies, response schemas, auth schemes, hosts, token hints, and validation still come from canonical Russian Swagger.

Validation

Use python3 scripts/wb_api.py validate after changing Swagger or helper code. Validation checks:

Russian and English specs parse successfully
duplicate method/path pairs
unresolved local $ref values
missing titles or operation tags
Russian/English mirror file and operation coverage
structural drift warnings between Russian canonical specs and the English overlay
optional generated markdown freshness with --generated

Structural drift in the English overlay is reported as a warning because Russian Swagger remains canonical for API behavior.

Source

Swagger files are local snapshots of Wildberries Seller API documentation. Replace or regenerate the Swagger snapshots when Wildberries updates the API. Derived markdown references are intentionally not kept in this repository; generate them on demand with scripts/wb_api.py generate.

FAQ

Common questions

Discussion