This release adds 1 notable feature for engineering teams evaluating rollout.
✓ No known CVEs patched in this version
Summary
AI summaryFixed model picker display issues across plans and resolved inference profile prefix errors for GovCloud regions.
Changes in this release
| Type | Severity | Summary | CVE |
|---|---|---|---|
| Feature | Low |
Adds `wheelScrollAccelerationEnabled` setting to disable mouse-wheel scroll acceleration in fullscreen mode Adds `wheelScrollAccelerationEnabled` setting to disable mouse-wheel scroll acceleration in fullscreen mode Source: llm_adapter@2026-06-12 Confidence: high |
— |
| Feature | Low |
Adds usage attribution to VSCode Account & usage dialog showing cache misses, long context, subagents, and per‑skill/agent/plugin/MCP breakdowns over 24h or 7d Adds usage attribution to VSCode Account & usage dialog showing cache misses, long context, subagents, and per‑skill/agent/plugin/MCP breakdowns over 24h or 7d Source: granite4.1:30b@2026-06-12-audit Confidence: low |
— |
| Bugfix | High |
Fixes Bedrock GovCloud regions deriving wrong inference profile prefix, causing 400 errors on model IDs Fixes Bedrock GovCloud regions deriving wrong inference profile prefix, causing 400 errors on model IDs Source: llm_adapter@2026-06-12 Confidence: high |
— |
| Bugfix | Medium |
Fixes `/model` picker hiding the model family that Default resolves to across plans Fixes `/model` picker hiding the model family that Default resolves to across plans Source: llm_adapter@2026-06-12 Confidence: high |
— |
| Bugfix | Medium |
Fixes `/model` picker showing hardcoded Sonnet version label when `ANTHROPIC_DEFAULT_SONNET_MODEL` pins a different Sonnet Fixes `/model` picker showing hardcoded Sonnet version label when `ANTHROPIC_DEFAULT_SONNET_MODEL` pins a different Sonnet Source: llm_adapter@2026-06-12 Confidence: high |
— |
| Bugfix | Medium |
Fixes "Fable 5 is now consuming usage credits" banner incorrectly showing for enterprise accounts with usage‑based billing Fixes "Fable 5 is now consuming usage credits" banner incorrectly showing for enterprise accounts with usage‑based billing Source: llm_adapter@2026-06-12 Confidence: high |
— |
| Bugfix | Medium |
Fixes background sessions inheriting another session's `ANTHROPIC_*` provider env variables Fixes background sessions inheriting another session's `ANTHROPIC_*` provider env variables Source: llm_adapter@2026-06-12 Confidence: high |
— |
| Bugfix | Medium |
Fixes 1‑2 second pause when exiting Claude Code shortly after interrupting a shell command on macOS and Linux Fixes 1‑2 second pause when exiting Claude Code shortly after interrupting a shell command on macOS and Linux Source: llm_adapter@2026-06-12 Confidence: high |
— |
| Bugfix | Medium |
Fixes incorrect git commit co‑author attribution showing wrong model name for some models Fixes incorrect git commit co‑author attribution showing wrong model name for some models Source: llm_adapter@2026-06-12 Confidence: high |
— |
| Bugfix | Medium |
Fixes `/advisor` dialog pre‑selecting a saved advisor model blocked by the `availableModels` allowlist Fixes `/advisor` dialog pre‑selecting a saved advisor model blocked by the `availableModels` allowlist Source: llm_adapter@2026-06-12 Confidence: high |
— |
| Bugfix | Medium |
Fixes skill hot‑reload re‑sending entire skill listing when a single skill changes; now only changed skills are announced Fixes skill hot‑reload re‑sending entire skill listing when a single skill changes; now only changed skills are announced Source: llm_adapter@2026-06-12 Confidence: high |
— |
| Bugfix | Medium |
Fixes Workflow tool `agent()` subagents missing per-agent attribution headers Fixes Workflow tool `agent()` subagents missing per-agent attribution headers Source: granite4.1:30b@2026-06-12-audit Confidence: low |
— |
| Bugfix | Medium |
Fixes pre‑warmed background workers failing with "Could not resolve authentication method" after idle period Fixes pre‑warmed background workers failing with "Could not resolve authentication method" after idle period Source: granite4.1:30b@2026-06-12-audit Confidence: low |
— |
Full changelog
What's changed
- Added
wheelScrollAccelerationEnabledsetting to disable mouse-wheel scroll acceleration in fullscreen mode - Fixed the
/modelpicker hiding the model family that Default resolves to — Opus now appears as its own row on Max/Team Premium/Enterprise plans, Sonnet on Pro/Team plans, and Opus on pay-as-you-go API accounts - Fixed
/modelpicker showing a hardcoded Sonnet version label whenANTHROPIC_DEFAULT_SONNET_MODELpins a different Sonnet - Fixed the "Fable 5 is now consuming usage credits" banner incorrectly showing for enterprise accounts with usage-based billing
- Fixed Bedrock GovCloud regions (
us-gov-*) deriving the wrong inference profile prefix (globalinstead ofus-gov), causing 400 errors on derived model IDs - Fixed background sessions inheriting another session's
ANTHROPIC_*provider env (gateway URL, custom headers,/modelaliases) from the shell that started the background daemon - Fixed a 1-2 second pause when exiting Claude Code shortly after a shell command was interrupted or killed on macOS and Linux
- Fixed git commit co-author attribution showing an incorrect model name for some models
- Fixed the
/advisordialog pre-selecting a saved advisor model that is blocked by theavailableModelsallowlist - Fixed skill hot-reload re-sending the entire skill listing when a single skill changed; only changed skills are now re-announced
- Fixed Workflow tool
agent()subagents missing per-agent attribution headers - [VSCode] Added usage attribution to the Account & usage dialog (
/usage) showing cache misses, long context, subagents, and per-skill/agent/plugin/MCP breakdowns over the last 24h or 7d - Fixed pre-warmed background workers failing with "Could not resolve authentication method" when claimed after sitting idle
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About claude-code
All releases →Related context
Earlier breaking changes
- v2.1.215 Claude no longer automatically runs /verify and /code-review skills
- v2.1.160 Renames dynamic‑workflow trigger keyword from `workflow` to `ultracode`; `workflow` no longer triggers a run
- v2.1.160 Deprecates and removes the `CLAUDE_CODE_OPUS_4_6_FAST_MODE_OVERRIDE` environment variable; it is now a no‑op
- v2.1.147 Renames /simplify to /code-review; removes cleanup-and-fix behavior.
Beta — feedback welcome: [email protected]