Fetching from the wire…
01
02
03
04
05
06
07
08
Public story · 2026-07-01 · high
The June 30 release runs on Bedrock and Google Cloud, letting IT run Claude Code centrally instead of trusting per-developer logins.
Why now: The gateway landed June 30 and appeared in this week's briefing, timing that matters for any stalled Claude Code pilot.
Anthropic shipped a self-hosted control plane for Claude Code on June 30, adding corporate SSO and spend caps, per Releasebot.
For any shop that couldn't let developers run Claude Code with personal logins, this closes the gap. IT can now enforce policy centrally and track cost per user instead of trusting whatever each engineer set up alone.
The Claude Apps Gateway runs on Amazon Bedrock or Google Cloud and layers in role-based access alongside SSO and spend caps.
It's a deployment layer, not a coding feature. It doesn't touch how Claude writes code or orchestrates subagents. It touches who gets to run it, under what policy, and how much they can spend.
My read: this is the release that gets Claude Code past security review in regulated shops, not the next model upgrade. The blocker was never capability, it was control.
The gateway landed June 30, folded into this week's briefing. Any team with a stalled Claude Code pilot should check whether it clears their blocker now.
Each link below shares sources, entities, or timing with this story.
Anthropic's own docs confirm prompts, model responses, tool results (which include code Claude reads) and session transcripts still travel to and are retained by Anthropic. Inference can't be routed through Amazon Bedrock, Google Cloud's Agent Platform, Microsoft Foundry, or a...
The v2.1.205 release turned /doctor into a full setup audit that flags unused skills, MCP, and plugins against their context cost, deduplicates local vs checked-in CLAUDE.md, and flags slow hooks (Releasebot). A typical 5-server, 58-tool MCP setup burns ~55k tokens before your...
The same man whose framework a model regression destroyed also published the most aggressive prediction of the week, and the tension between those two facts is the whole argument. "The Shape of Things to Come, Part 1: The Continuous Thunderdome" argues traditional CI/CD collap...
Two things in one release, and only one of them is in the security section. CLAUDE_CODE_SUBAGENT_MODEL used to override the model for every subagent. In 2.1.251 it sets the default instead. An agent definition's model: frontmatter now wins, and so does an explicit per-spawn mo...
AWS's August 21 post stages agent tool governance as Connect, Control, Catalog and Harden, from one SSO-backed MCP endpoint for a 1-20 user pilot through identity-aware authorization with PII redaction and self-service tool publishing at 100+ users. It supports Cognito-backed...
Releasebot lists migration of settings, MCP servers, plugins, sessions, commands, and project-scoped memories, plus stabilized multi-agent V2 with per-subagent models, reasoning levels and concurrency, and Bedrock login with GPT-5.6 Sol as default. The migration path is the no...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.