Book · $19.99

The Agent Workbench, Claude Code edition

Find out in two weeks whether Claude Code is paying off for your team

Plan the work, check the evidence, make a better decision. A practical companion for tech leads and engineering managers: a two-week way to assess how Claude Code is used and controlled, and what stronger evidence you need before you claim it pays off.

Who it’s for

• Tech leads and engineering managers evaluating Claude Code for a team.
• Developers who want a repeatable plan, check and handoff routine.

What you get

• Set up in an afternoon The plan, check and handoff routine, and approval hooks with their limits stated.
• Run the readout What a team transcript readout shows: sessions, failed tool calls, test runs, guardrail refusals.
• Read it with the playbook What each signal usually means, and what to change.
• Write the memo and decide A one-page manager memo with recommendations, a decision block and what the numbers can’t show.
• Prove it on your own code Keep context, tokens, plan limits and API charges apart, and set a baseline before claiming time saved.

Buy the PDF on Gumroad · $19.99

Terms

Format

One downloadable PDF, 29 pages including covers, November 2026, delivered by Gumroad right after checkout.

Scope

The PDF only. The code kits, installers, skill folders and templates the books describe are separate and not included.

Honest notes

• Independent publication by Tim Urista. Not affiliated with or endorsed by Anthropic. Claude and Claude Code are trademarks of Anthropic.
• Example projects and team numbers are synthetic and labeled as such. They are not customer results.
• No security, savings or productivity guarantees.