CCAR-F scenarios › Developer Productivity with Claude · Question 39 of 60

Scenario 4 · Developer Productivity with Claude

Your team is building internal developer tools on the Claude Agent SDK. The agent helps engineers find their way around unfamiliar or legacy codebases, generates boilerplate, and automates repetitive chores. It works through the built-in Read, Write, Bash, Grep, and Glob tools, plus connections to Model Context Protocol (MCP) servers.

Agentic Architecture & Orchestration

An engineer spent a long session having the agent analyze a legacy authentication module. They now want to compare two refactoring approaches, an adapter layer and an incremental rewrite, both starting from that analysis, without redoing the exploration and without either approach's experiments influencing the other. What should they do?

Show answer and explanation

Answer: B — Use fork_session to create two independent branches from the analysis session, and develop one refactoring approach in each branch.

fork_session creates independent branches from a shared analysis baseline, which is the case section 6 describes for comparing refactoring approaches. (a) and (d) leave both approaches in one context, so the first one shapes the second. Telling the agent to "disregard" earlier work doesn't remove it from context. (c) throws away the analysis already done and pays for exploration twice.