CLAUDE LABJP
BILLING — Day two after the Jun 15 change: Agent SDK, headless runs, GitHub Actions, and third-party agents now bill against separate monthly credits ($20/$100/$200) at full API rates with no rollover, making first-day cost measurements the basis for any reworkREGULATED — TCS partnered with Anthropic to bring Claude to banks, airlines, and other regulated industries, while DXC integrates Claude into the core systems those sectors rely onRETIRED — Sonnet 4 and Opus 4 left the API on Jun 15; confirm via your logs that scripts referencing them have moved to the latest generation such as Opus 4.8EXPORT — Claude Fable 5 and Mythos 5 remain suspended under a US export-control directive (since Jun 12); Anthropic says it is working to restore accessSAFE — Only the two new Mythos-class models are affected; every other model including Opus 4.8 keeps running normallySUBAGENTS — Claude Code sub-agents can spawn their own sub-agents up to five levels deep, widening the design space for multi-stage delegationBILLING — Day two after the Jun 15 change: Agent SDK, headless runs, GitHub Actions, and third-party agents now bill against separate monthly credits ($20/$100/$200) at full API rates with no rollover, making first-day cost measurements the basis for any reworkREGULATED — TCS partnered with Anthropic to bring Claude to banks, airlines, and other regulated industries, while DXC integrates Claude into the core systems those sectors rely onRETIRED — Sonnet 4 and Opus 4 left the API on Jun 15; confirm via your logs that scripts referencing them have moved to the latest generation such as Opus 4.8EXPORT — Claude Fable 5 and Mythos 5 remain suspended under a US export-control directive (since Jun 12); Anthropic says it is working to restore accessSAFE — Only the two new Mythos-class models are affected; every other model including Opus 4.8 keeps running normallySUBAGENTS — Claude Code sub-agents can spawn their own sub-agents up to five levels deep, widening the design space for multi-stage delegation
TAG

Rollback

2 articles
Back to all tags
Related:
Claude Code1Refactoring1Git1Observability1claude-agent-sdk1progressive-delivery1canary1feature-flags1production1
Claude Code/2026-06-16Advanced

Keeping Large Claude Code Refactors Revertible One Commit at a Time — Field Notes on Checkpoints and Rollback Detection

Hand a big refactor to Claude Code and the speed hides a real cost: review-proof, oversized diffs. Here are the field notes I actually run — declaring checkpoints in a manifest, enforcing commit granularity with a pre-push hook, and tying rollback calls to observability.

API & SDK/2026-04-22Advanced

Implementing Progressive Delivery with the Claude Agent SDK: Canary, Feature Flags, and Automatic Rollback Patterns for Production

Production-grade patterns for safely rolling out AI agents built with the Claude Agent SDK. Combines canary traffic splitting, feature flags, and SLO-driven automatic rollback with runnable TypeScript/Hono implementation code.