Fetching from the wire…
01
02
03
04
05
06
07
08
Public story · 2026-09-02 · high
The update adds a 1M-token context window at $10/$50 per million tokens, plus a way to force one model onto every subagent.
Why now: Anthropic pushed the update on September 1, with a same-day follow-up patch fixing a macOS 12 launch bug hours later.
Anthropic released claude-code v2.1.257 on September 1, making claude-fable-5-1 the default Fable model with a 1M-token context window. Whole repos or long transcripts now fit in one pass, cutting the need to chunk work across multiple calls.
Input tokens cost $10 per million, output tokens $50 per million, and cached reads $0.25 per million, per the claude-code changelog.
The same release adds CLAUDE_CODE_SUBAGENT_MODEL_FORCE, an environment variable that overrides the model for every subagent in a session. It ignores whatever a per-spawn call or an agent definition file requests. Anyone running a cheap model for routine subagent work and a stronger one for judgment calls loses that split the moment the variable is set.
There's also a session-scoped s option added to /effort, letting reasoning effort get set once per session instead of per call.
A follow-up release, v2.1.258, went out the same day at 22:33 UTC, fixing a macOS 12 launch regression that 2.1.255 had introduced.
Each link below shares sources, entities, or timing with this story.
The release also adds a command that migrates Python projects off the old anthropic 0.x SDK and stops WebFetch from serving stale content all session.
Two of the six hid commands from the approval dialog itself, using tab padding, invisible Unicode, and split Bash strings.
Anthropic released Claude Fable 5.1 on September 1. Claude Code v2.1.257 made it the default Fable model at 17:53 UTC that day, with a 1M-token context window, $10 per million input tokens, $50 per million output, and $0.25 per million on cache reads (claude-code CHANGELOG). B...
Server-managed settings that terminate sandbox TLS, route sandbox traffic through your proxy, inject credentials, or weaken isolation all need approval before applying. ANTHROPIC_CUSTOM_HEADERS from managed or project settings needs approval when it sets a credential, org/tena...
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...
Claude Code 2.1.251 fixes checks that auto-approved commands assigning an arithmetic expression to an integer shell variable, naming OPTIND=1/0 and RANDOM=2+2. Both now prompt. The same release changes how Bash command output files are created and read back in the sandbox so a...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.