Don't Accept an Agent's Numbers and Citations As-Is — A Verification Gate Built on a Dedicated Auditor Subagent
A design that verifies every number and citation in an agent-generated summary using a separate subagent before accepting it — with working TypeScript for deterministic recomputation and fail-closed source matching.
When Claude API Extracts the Wrong Value With Full Confidence — Designing the Verification Layer
When you extract invoices or contracts with Claude API, the scariest failure isn't an exception — it's plausible-but-wrong JSON. Here is how I build a verification layer that catches silent extraction errors with schema checks, arithmetic reconciliation, and dual-extraction agreement, in TypeScript.
Stopping Claude Code from Quitting Mid-Task — Prompt Patterns and Verification Loops
Claude Code says "Done!" — but the build is broken and only half the files were touched. Here are the prompt patterns and automated verification loops I now use in production to keep agents running until the job is actually finished.
Letting Claude Code Handle Dependency Upgrades — A Verification Loop That Catches Breaking Changes
Once a solo project carries 30+ dependencies, the monthly upgrade run becomes a dreaded ritual. This article maps out where Claude Code can take the wheel, where humans must keep judgment, and the verification loop I use to keep breaking changes out of production.