explaingit

frankhuy/fishkeeper

14PythonAudience · developerComplexity · 3/5Setup · moderate

TLDR

A Python toolkit that connects AI coding assistants to Xianyu's seller API, letting you manage products and orders on Alibaba's second-hand marketplace by describing tasks in plain language.

Mindmap

mindmap
  root((fishkeeper))
    What it does
      Automate Xianyu store
      AI agent skill
      Bulk operations
    Product management
      Create and edit items
      List and delist
      Delete in bulk
    Order management
      Query orders
      Mark as shipped
      Change prices
    AI integrations
      Claude Code
      OpenAI Codex CLI
      Cursor
      MCP compatible
    Auth
      MD5 signature signing
      App key and secret
Click or tap to explore — scroll the page freely

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

Things people build with this

USE CASE 1

Let an AI assistant list or delist products on your Xianyu store from a plain-language description.

USE CASE 2

Batch-ship multiple orders at once using the multi-agent parallel API calls.

USE CASE 3

Query card key inventory for digital goods and look up shipping options automatically.

Tech stack

PythonXianyu APIMCPMD5

Getting it running

Difficulty · moderate Time to first run · 30min

Requires a Xianyu open platform account with an app key and app secret.

License not specified in the explanation.

In plain English

Fishkeeper is a Python toolkit for managing a store on Xianyu, a Chinese second-hand marketplace operated by Alibaba. The project is written primarily in Chinese and is aimed at sellers or developers who want to automate store operations through Xianyu's open platform API rather than managing everything manually through the app. The toolkit is structured as a skill that can be loaded by AI agent frameworks. The README describes how to install it into several different AI coding environments: Hermes Agent (which is the primary target), Claude Code, OpenAI Codex CLI, Cursor, and any system that supports the Model Context Protocol. In each case the idea is the same: once the skill files are in place, the AI assistant can call the Xianyu API on behalf of the user when asked to do store-related tasks. The core functionality covers the full lifecycle of selling on Xianyu. On the product side it handles creating, editing, listing, delisting, and deleting items, including bulk operations. On the order side it handles querying orders, marking items as shipped, changing prices, and looking up card keys (which are used for digital goods). It also supports querying shipping companies and category or attribute lookups for setting up product listings correctly. Authentication to the Xianyu API uses MD5 signature signing, which is a common approach for Chinese open platform APIs. API credentials (an app key and an app secret) can be provided as environment variables or in a configuration file. The toolkit covers 19 API endpoints plus 3 webhook callback types for push notifications. The multi-agent orchestrator included in the project allows running several API calls in parallel, which is useful for operations like syncing inventory across a large catalog or batch-shipping multiple orders at once. The README is written for developers and assumes familiarity with Python and AI agent tooling.

Copy-paste prompts

Prompt 1
I have fishkeeper installed as a Claude Code skill. Write me a prompt to tell Claude to list 5 new items on my Xianyu store with specific titles and prices.
Prompt 2
Show me how to configure fishkeeper with my Xianyu app key and secret using environment variables, then have it query all pending orders.
Prompt 3
Using fishkeeper in Claude Code, how do I bulk-ship all orders placed in the last 24 hours?
Open on GitHub → Explain another repo

← frankhuy on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.