It's easy to treat tools as an implementation detail, a list engineering assembles once the requirements are clear. That gets it backwards. The tool list is the requirements. An agent can do exactly what its tools allow and nothing else, so deciding what tools exist is deciding what the product is.

Decision	Who Usually Makes It	Who Should
Which tools exist at all	Engineering, by default	You, it defines the product's scope
What each tool is called and described	Engineering	You, with engineering, it's user-facing in effect
How granular each tool is	Engineering	Together, it's a capability and reliability trade-off
What a tool returns when it fails	Engineering	You, it determines what the user eventually sees
Which tools require approval	Sometimes nobody	You, it's a risk decision
How the tool is implemented	Engineering	Engineering, genuinely

Agentic AI and MCP for Product Managers

Module 1What 'Agentic' Actually Means

Module 2How an Agent Actually Works

Module 3MCP: What It Is and Why You Should Care

Module 4Designing the Tool Layer

Module 5Permissions and the Human in the Loop

Module 6Evaluating Agents

Module 7Cost, Latency and Unpredictability

Module 8Designing What the User Sees

Module 9Risk and Blast Radius

Module 10When an Agent Is the Wrong Answer

Module 11Final Project: Design One Real Agent End to End

Module 12Appendix: The Prompt Library

Why Tool Design Is Product Design