# Agent Prompt Hub · agent.ifq.ai

> A bilingual Agent Prompt hub for e-commerce business experts getting started with AI agents. The site teaches business users to write useful prompts with a simple framework: goal, input, action, deliverable. Content pages and AI-readable resources require sign-in when Supabase auth is configured.

## Primary content

- [Home](https://agent.ifq.ai/en): Overview of the Agent Prompt framework and real e-commerce Agent cases.
- [Prompt Generator](https://agent.ifq.ai/en/generator): Fill-in form that produces a copy-ready Agent Prompt in Chinese or English, plus a revision prompt for weak outputs.
- [Onboarding Guide](https://agent.ifq.ai/en/guide): Plain-language guide to goal, input, action, deliverable, evidence, and iteration.
- [Writing Framework](https://agent.ifq.ai/en/cases): The four-part Agent Prompt framework with copy-ready starters.
- [Agent Task Cases](https://agent.ifq.ai/en/agent-cases): E-commerce Agent task cases for sales briefs, order follow-up, support analysis, product launch checks, inventory, procurement, ads, members, risk, live commerce, category reviews, and review insights.
- [Prompt Recipes](https://agent.ifq.ai/en/cookbook): Short copy-ready prompts for common business tasks.
- [Tips](https://agent.ifq.ai/en/tips): Practical prompt habits and common prompt mistakes.
- [FAQ](https://agent.ifq.ai/en/faq): Common Agent Prompt questions for business teams getting started with agents.

## Key model

Good Agent Prompts clearly state:

1. Goal: who uses the result and what problem it solves.
2. Input: what tables, documents, chats, or context the agent will receive.
3. Action: what the agent should inspect, classify, analyze, or decide.
4. Deliverable: the final format, such as a brief, table, action plan, or JSON.

## AI-readable resources

- LLM summary: https://agent.ifq.ai/llms.txt
- Full LLM index: https://agent.ifq.ai/llms-full.txt
- Agent guide: https://agent.ifq.ai/agent.md
- Compatibility agent guide: https://agent.ifq.ai/agents.md
- Sitemap: https://agent.ifq.ai/sitemap.xml

## Use policy

Content and prompts are MIT-licensed. Preserve the business-expert, AI-onboarding framing when quoting.