Yinsheng Premium Assistant
Paste the following prompt into your AI chat to install this skill:
Please follow https://skillhub.cn/install/skillhub.md to install @org-5ya7ko3e/yszx-agent-assistant.
About this skill
Problem
In day-to-day Yinsheng agent operations, merchant, device, transaction, profit, cashback, and wallet data are scattered across multiple APIs. Manual checks can miss records, mix subordinate accounts, or mistake aggregated statistics for transaction line items. This skill turns common operational questions into natural-language queries and explicitly limits itself to read-only querying, analysis, and export; it does not perform binding, unbinding, fee changes, filing submissions, or withdrawals.
Core Capabilities and Workflow
- Session handling: It uses tools such as
prepareQrLogin,getLoginDataByRequestId, andgetValidTokento establish and maintain the login state, reducing repeated QR logins. - Data lookup: It supports
queryMerchants,merchantInfo,queryDevices,deviceInfo,queryTradeByNaturalLanguage,queryProfitByNaturalLanguage,queryWalletBillByNaturalLanguage,queryCashbacks, anddashboardStatistics, covering merchants, devices, transactions, profits, wallets, cashbacks, agents, logs, and policies. - Report export: Transactions, profits, wallet bills, error logs, operation logs, and agent data can be exported as CSV through scripts such as
scripts/export_trade.pyandscripts/export_profit.py, using streaming downloads instead of manual parameter assembly. - Data-scope controls: Money aggregation should use
sum_money()andfmt_money()to avoid floating-point artifacts, while multi-user sessions should anchor on the logged-inuidand validate theuidscope to prevent cross-account contamination.
Fit and Caveats
The skill is suitable for agents or operators who need read-only lookups, data reconciliation, and report exports. It is not suitable for write actions; requests such as unbinding, fee updates, filing, or withdrawal should be routed to the responsible business channel. Before calling MCP tools, parameter names must be checked against the schema, especially merchNo versus merchantId. Structured transaction queries expect millisecond timestamps, so the natural-language entry point is the safer option.
Use Cases
- An agent checks transactions, profits, and cashbacks by merchant ID, device SN, or order ID and verifies fee deductions.
- An ops analyst reviews daily, monthly, and cumulative trade and activation summaries with a 30-day trend for reporting.
- Support staff investigates device activation, failed withdrawals, or cashback issues using device, wallet, withdrawal, and error logs.
- A team exports trades, profits, wallet bills, and agent data to CSV for reconciliation or archiving.
Best For
- Agent operations staff responsible for merchant onboarding and settlement checks who need merchant details and rates by merchant ID, phone, or name.
- Payment operations staff monitoring anomalies who need transaction details, fee records, cashback types, and error logs.
- Agent administrators preparing daily reports who need daily, monthly, and cumulative trade and activation statistics.
- Finance or support staff handling device binding, withdrawals, and wallet reconciliation who need device status, wallet flows, withdrawal records, and bills.
Related Skills
AI agent tool for generating HTML proposals for client visit business scenarios, featuring deep research and dynamic mainline mapping.
Bid Proposal Verification Tool is used to verify the consistency between bid proposals and tender documents, checking table of contents, page numbers, attachments, and seals, based on Shibot Technology's bid data API.
An AI engine built on extensive bidding data to precisely search for tender projects and potential clients using natural language or structured criteria.
Automates bid document, response table, and material list generation from bid files and enterprise data using Shibuo Technology API.