Sample Deliverables
Below is the directory structure and content summary of a standard Agent delivery package (anonymized)
Package Directory Structure
README.md — Project overview & quick start
agent-spec.json — Agent goals, capability boundaries, I/O definitions
system-prompt.md — Full system prompt text
workflow.md — Workflow steps, decision logic, exception handling
test-cases/ — 26+ test cases (normal / edge / error / high-risk / adversarial)
platform-adapters/ — Claude Code / Codex / OpenClaw / n8n deployment guides
user-guide.md — Illustrated client user manual
Agent Spec Example (Summary)
Agent Name
Customer FAQ Auto-Reply Agent
Business Goal
Auto-handle 80% of common questions, escalate complex ones to humans
Target User
E-commerce customer service team
Risk Boundary
No refund amount promises, no order modifications, no formal external communications
System Prompt Example (Excerpt)
You are a professional customer service assistant. Your responsibilities:
1) Answer common questions based on the FAQ knowledge base
2) Identify complex issues requiring human intervention and classify them
3) Never promise refund amounts or modify orders
4) When uncertain, reply 'I need to confirm this and get back to you'
Workflow Steps Summary
011. Receive customer message → Extract key information
022. Match FAQ knowledge base → Confidence > 0.85 auto-reply
033. Confidence < 0.85 → Classify issue type (logistics/refund/specs/other)
044. Logistics/Specs → Query system and reply
055. Refund → Flag for human, generate ticket summary
066. Human review checkpoints: Refund confirmation, abnormal complaints
Test Case Examples