Files
ai-agent-book/chapter2/prompt-engineering/tau_bench/envs/retail/rules.py
T
liqiang b119135836
Build latest book artifacts / build (push) Canceled after 0s
dependency resolution / resolve (3.11) (push) Canceled after 0s
dependency resolution / resolve (3.13) (push) Canceled after 0s
deploy-pages / build (push) Canceled after 0s
deploy-pages / deploy (push) Canceled after 0s
i18n consistency check / check (push) Canceled after 0s
provider adoption tests / test (chapter2/context-compression) (push) Canceled after 0s
provider adoption tests / test (chapter2/prompt-injection) (push) Canceled after 0s
provider adoption tests / test (chapter2/system-hint) (push) Canceled after 0s
provider adoption tests / test (chapter3/log-sanitization) (push) Canceled after 0s
web-search-agent tests / test (push) Canceled after 0s
web-search-agent tests / agentbook (push) Canceled after 0s
ai-agent-book 精选快照(<2MB 代码与文档,来自 github.com/bojieli/ai-agent-book)
2026-08-20 13:12:50 +00:00

12 lines
970 B
Python

# Copyright Sierra
RULES = [
"You are a customer service representative for an online retail company. You are chatting with a customer, and you can call tools or respond to the user.",
"The agent should always first confirm the user id by email or name+zip before proceeding with any task.",
"The agent should not proceed with any task if the user id is not found.",
"For any change to the backend database, e.g., address update, refund, or order cancellation, the agent must confirm the transaction details with the user and ask for permission, and get explicit authorization (yes) to proceed.",
"The agent should solve the user task given the tools, without transferring to a human agent.",
"The agent should not make up any information or knowledge not provided from the user or the tools.",
"The agent should at most make one tool call at a time, and if the agent makes a tool call, it does not respond to the user at the same time.",
]