Wildberries Seller API skill
Wildberries Seller API - Agent Skill
A Wildberries Seller API reference skill on local Swagger files with search in Russian and English
- 7 ★ on the repository
- Last commit
Extended access
Why this level
- The agent runs the Python script wb_api.py to search the WB Swagger specs
- The script works with local files and makes no network calls
Install
- Works in
- Services
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/wildberries-seller-api-agent-skillDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The agent runs the Python script wb_api.py to search the WB Swagger specs
- The script works with local files and makes no network calls
This is third-party code. Review the repository files before installing.
What it does
The repository holds Russian WB Swagger specs as the canonical source and an English mirror for search and labels. The wb_api.py script shows a section map, searches endpoints, expands request and response schemas, and prints auth and host details. It makes no network calls, so the agent does not need to load full specs into context.
Who it is for. For Wildberries integration developers working in Codex or Claude Code.
Good fit when
- You need an endpoint for cards, prices, stocks or orders
- You need the request and response schema of a method
- You need a Python API call with the correct host and header
Not a fit when
- You need a client that actually calls the API
- You need guaranteed fresh specs: snapshots are updated manually
Example request
Find the seller rating endpoint and show its response schemaLimitations
Swagger files are local snapshots that must be updated manually when the API changes. The English mirror is auxiliary and drift is reported as warnings. Requires Python with PyYAML.
How to disable. Remove the wildberries-api folder from your agent's skills directory.
FAQ
Which specs are canonical?
The Russian swagger/*.yaml files. The English swagger/en/*.yaml files are used for search and labels; parameters and schemas come from the Russian ones.
Discussion
No questions or reviews yet.
Related
All Wildberries toolsA Wildberries Seller API reference built on an OpenAPI snapshot the author syncs with the WB portal about monthly
Wildberries API skill with 261 methods
wb-api skill
An Agent Skill with a full map of the Wildberries API: 12 modules, 261 methods, reference files and a call script
Official Wildberries Seller API skill
Wildberries API
A skill for the official Wildberries Seller API with local OpenAPI schemas for 286 operations and a guarded api_call helper
Wildberries Seller API reference
wb-seller-api-reference
A Claude Code skill with a Wildberries Seller API reference: hosts, rate limits, errors and 18 endpoint families
A zero-dependency Agent Skill and CLI over the Wildberries API: products, orders, reviews, analytics, multiple stores
A conservative CLI and skill for the official Wildberries Seller API: token checks, ping, seller-info and read-only raw requests
README in short
The README explains installing the skill for Codex and Claude Code by cloning into the skills directory. wb_api.py supports validate, map, search, tag, detail, protocol and generate. Validation checks parsing, duplicates, unresolved refs and mirror drift. Derived markdown docs are generated on demand and not stored in the repo.
README badge
Original SKILL.md
--- name: wildberries-api description: Swagger-backed Wildberries seller API reference for WB marketplace integrations. Use when building or debugging code that calls Wildberries/WB seller API endpoints, choosing endpoints, checking request/response schemas, authentication headers, hosts, rate limits, or API capabilities. --- # Wildberries API Treat local Swagger as the only API source of truth. Russian files in `swagger/*.yaml` are canonical; English files in `swagger/en/*.yaml` are a validated translation overlay for English search and text labels. Use the helper script for compact Swagger-derived lookups; do not load full Swagger files unless the helper output is insufficient. ## Default Workflow The helper requires PyYAML. If it is missing, install dependencies with `python3 -m pip install -r requirements.txt`. 1. Start with the compact Swagger query script: - Get an overview: `python3 scripts/wb_api.py map` - Search endpoints in Russian or English: `python3 scripts/wb_api.py search "<query>" --limit 25` - Inspect an exact endpoint: `python3 scripts/wb_api.py detail <METHOD> <PATH>` - Expand request/response fields: `python3 scripts/wb_api.py detail <METHOD> <PATH> --schemas` - Inspect auth, hosts, token hints, ping, and status codes: `python3 scripts/wb_api.py protocol` - List observed hosts/categories only: `python3 scripts/wb_api.py hosts` 2. Open raw `swagger/*.yaml` only when the script output is insufficient for exact schemas, nested examples, enum values, or long endpoint descriptions. 3. If human-readable browse docs are needed, generate them on demand with `python3 scripts/wb_api.py generate`; generated output is ignored by git. 4. Validate parser coverage and the Russian/English mirror before relying on output: `python3 scripts/wb_api.py validate`. ## Answering Rules - Cite method, path, host, source Swagger file, auth scheme, parameters/body, and relevant response codes. - Do not invent global protocol details. For rate limits, token types, sandbox behavior, or status meaning, use endpoint descriptions, `x-token-types`, `servers`, `securitySchemes`, and responses from Swagger. - Prefer compact script output over loading complete Swagger files into context. - If the local Swagger snapshot does not contain what the user expects, say that directly and ask before checking live docs. ## Code Generation Checklist - Read credentials from env/config; never hard-code the token. - Send the token using the Swagger-declared `HeaderApiKey` scheme (`Authorization` header). - Use the host declared on the operation or inherited by its path/spec. - Implement pagination, polling, retries, and payload validation only when the endpoint schema or description calls for it.