CLAUDE LABJP
SUNSET — The legacy Workbench retires today, August 17: saved prompts, prompt versions, and evals become inaccessible after that, so exporting is a today-or-never jobAPI — The experimental prompt tool endpoints generate_prompt, improve_prompt, and templatize_prompt retire the same day and will return errors, so any script calling them needs switching over todayCONSOLE — The replacement Workbench is stateless: nothing is stored on Anthropic's servers, your draft stays in the browser, and any request can be exported as codeOSS — Claude for Open Source now grants six months of Claude Max 20x to qualifying maintainers, capped at 10,000 people, individual only, with no API credits and no auto-renewalQUOTA — The 50 percent weekly usage boost for Claude Code subscribers runs through August 19, two days outSELF-HOSTED — Self-hosted environments for Claude Code are in public beta, letting Team and Enterprise plans run sessions on their own infrastructure with internal network access and custom toolingSUNSET — The legacy Workbench retires today, August 17: saved prompts, prompt versions, and evals become inaccessible after that, so exporting is a today-or-never jobAPI — The experimental prompt tool endpoints generate_prompt, improve_prompt, and templatize_prompt retire the same day and will return errors, so any script calling them needs switching over todayCONSOLE — The replacement Workbench is stateless: nothing is stored on Anthropic's servers, your draft stays in the browser, and any request can be exported as codeOSS — Claude for Open Source now grants six months of Claude Max 20x to qualifying maintainers, capped at 10,000 people, individual only, with no API credits and no auto-renewalQUOTA — The 50 percent weekly usage boost for Claude Code subscribers runs through August 19, two days outSELF-HOSTED — Self-hosted environments for Claude Code are in public beta, letting Team and Enterprise plans run sessions on their own infrastructure with internal network access and custom tooling
TAG

Regression Detection

2 articles
Back to all tags
Related:
Claude Code1Refactoring1Contract Testing1Observability1Claude API1Opus1automation1prompt design1
Claude Code/2026-07-03Advanced

When a Claude Code Refactor Passes Every Test but Behaves Differently in Production — Catching Silent Contract Drift with a Behavior Diff Harness

Hand Claude Code a large refactor and your tests can stay green while production behavior quietly shifts. Here is how I record exception channels, log shape, init order, and return values as a signature, then diff them per commit to catch contract drift before it ships.

API & SDK/2026-06-26Advanced

When the Same Model Name Starts Behaving Differently: A Startup Canary for Unattended Pipelines

An in-place Opus upgrade can change your output, and an unattended publishing pipeline will never notice. Here is a lightweight startup canary that fingerprints behavior, catches drift, and halts the batch — with measured cost and latency.