Fetching from the wire…
Tools2026-09-23 · source-backed
The release fixes writes through a symlinked path being judged by their in-tree spelling, where acceptEdits, allow rules and auto mode could approve a write landing outside the tree. Auto mode also stops retrying forever: one denial when a safety check declines to review, backoff when it returns no answer, turn ends after ten consecutive denials. It also switches Pro and Team Standard defaults from Sonnet to Opus, adds CLAUDE_CODE_MAX_MCP_DESCRIPTION_LENGTH to override the 2,048-character cap, and fixes a finished subagent's report being lost when the parent compacts before reading it.
Each link below shares sources, entities, or timing with this story.
Within five days of each other, both Claude Code (v2.1.158, May 31) and Cursor (3.6, May 29) shipped remarkably similar architectures for autonomous agent execution. Both use a classifier subagent that reviews each pending action against conversation context and decides: allow...
Buried in the 2.1.233 notes. The harness is removing explicit planning scaffolding for the newer tier while keeping it for older models, betting these models track multi-step work internally and that a forced todo list burns tokens without improving completion. Claude Code Rel...
A GitHub repo cataloging Claude Code tips doesn't normally warrant a top story. But shanraisshan/claude-code-best-practice at 53.4K stars isn't a tips list anymore. It's the de facto reference for how an entire generation of developers is learning to work with AI coding agents...
If you're on Pro, Max, or Team, the permission prompt you've been hitting Enter on for a year goes away Friday. Anthropic confirmed auto mode becomes the default, replacing per-call approval with a classifier that inspects each tool call for irreversible, destructive, or out-o...
Tessl ran 880 evaluations across 9 models with and without agent skills. The result inverts what most teams assume about AI costs. Haiku 4.5, Anthropic's cheapest model at roughly $0.25 per million tokens, scored 84.3% when given a well-crafted agent skill. Opus 4.7, the most...
If you've used Claude Code for any serious session, you know the drill. Approve. Approve. Approve. Approve. You stop reading the prompts after the fifteenth one. That's the worst possible security outcome, way worse than a well-designed automated check. Anthropic launched auto...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.