Giving an agent the right tools (and only those)

When you give an agent tools (web search, file edit, send email), each tool is a way it can mess up. Less is more.

Some examples

• Start with read-only tools (search, fetch) before write tools (send, delete).
• Never give a hobby agent tools that cost money without a hard limit.
• Log every tool call so you can audit later.

Try it!

Build an agent that summarizes your inbox. Give it READ access only — no reply, no delete.

Understanding "Giving an agent the right tools (and only those)" in practice: AI agents don't just answer questions — they can do things, like looking things up, writing files, or talking to apps. Agents are only as safe as the tools they can call — pick the smallest set that works — and knowing how to apply this gives you a concrete advantage.

• Design clear agent goals before adding tools
• Define permissions and scope before deploying any agent
• Build in human-approval checkpoints for high-stakes actions
• Understand when to use an agent vs. a simple chat prompt

1. 1Design an agent spec: goal, tools, permissions, stop condition
2. 2Run a simple web-search agent in a sandbox environment
3. 3Instrument an existing workflow to identify where an agent could save time