About this skill

Problem addressed

When users ask about COFCO Futures profile, account opening, trading calendars, delivery memos, account/position/order/transaction queries, or simulated order placement and cancellation, a generic model may rely on stale knowledge and invent financial details.
This skill gives the agent a clear contract: prefer live official data, degrade safely, and keep compliance boundaries explicit.

How it works

The agent follows SKILL.md and the MCP server definition in skill.json, calling tools such as get_company_profile over MCP / JSON-RPC 2.0.
User intent is routed to the matching tool or embedded skill:
get_company_profile for company overview, business qualifications, and exchange memberships.
delivery-inquiry for trading calendar and trading-content questions.
delivery-memo-inquiry for delivery memos and buyer/seller delivery details.
account-opening for onboarding and account manager contact.
collect_customer_info for callback requests, with explicit consent handling before collecting phone numbers.
futures-trading for contract info, account/position/order/transaction queries, and simulated order or cancel actions.
Answering priority is: MCP live data, then static documentation, then a retry message; sample data is only for format reference.

Boundaries

It does not provide specific investment advice, return promises, or market-direction predictions.
For real-time quotes or non-COFCO Futures services, it should direct users to official channels.
The tone is professional, steady, and objective, avoiding promotional language.

Screenshots

Use Cases

A customer service agent answers questions about COFCO Futures company profile, qualifications, exchange memberships, and branches.
During onboarding, the agent captures account-opening intent, requests callback consent, and explains the phone number is used by an account manager.
For delivery operations, the agent queries trading calendars, trading content, or buyer/seller delivery memo details by date and instrument.
In simulated trading, the agent checks contracts, account, positions, orders, executions, and handles order placement or cancellation.

Best For

Engineers building financial-domain chat agents need company overview, trading/delivery queries, and simulated trading capabilities.
Product managers designing customer service flows need intent routing to account opening, callback, delivery, and trading sub-skills.
Test engineers validating financial bots need to check contracts, accounts, positions, orders, executions, and order workflows.
Customer operations advisors need compliant callback collection and handoff to account managers.

Related Skills

Cold Shower Gate

@user_2da54b45/cold-shower

Switch AI from a polite executor into a skeptical thinking partner, using assumption checks, pre-mortems, and evidence gates to expose weak ideas early.

Professional 2026.08.29 ↓ 280