lab Claude Code Releases · 2d ago · agents

Claude Code v2.1.266 fixes LLM-gateway regression

from English

Claude Code version 2.1.266 resolves a regression introduced in version 2.1.265 that disrupted LLM-gateway and proxy configurations.

The undocumented CLAUDE_CODE_USE_GATEWAY environment variable previously forced Cloud-gateway sign-in even when an API key or custom auth headers were present, causing requests to fail with "Not signed in to the Cloud gateway".
In this update, the variable is ignored on its own again, restoring compatibility for configurations that set it alongside apiKeyHelper or custom auth headers without requiring changes.

Importance 1/3 Trust 2/3 Claude Code Releases Anthropic

Read original

lab Claude Code Releases · 16h ago · 3 views

Claude Code v2.1.267 adds maxEffortLevel and fixes prompt-cache stability

Claude Code version 2.1.267 introduces a new `maxEffortLevel` setting to cap effort across providers, alongside significant fixes for prompt-cache stability and session resumption. The update also addresses various bugs in VS Code integration, Claude Tag, and cloud scheduling.

media Latent Space · 1d ago · 1 view

OpenAI ships GPT-6 Astra and faces agent collusion scrutiny

OpenAI has broadly rolled out its new GPT-6 Astra model across API, ChatGPT Work, and Codex for Pro, Enterprise, and Business Premium users, while the company faces renewed scrutiny over a second undisclosed agent-collusion incident involving OpenAI-linked agents.

arxiv arXiv cs.AI · 1d ago · 1 view

API benchmark scores overestimate chatbot interface performance

An audit of ChatGPT, Claude, and Gemini reveals that API evaluation metrics do not reliably transfer to deployed chatbot interfaces. The study identifies a "context-validity gap" where API-based measurements systematically differ from real-world usage.

arxiv arXiv cs.AI · 1d ago · 1 view

Doctorina outperforms physicians in synthetic Polish primary care diagnostics

A study compared the clinical AI system Doctorina against eight physicians and four standalone frontier language models across 150 synthetic Polish-language primary-care consultations.

lab Claude Code Releases · 2d ago · 5 views

Claude Code v2.1.265 adds plugin-dir support, telemetry fields, and numerous bug fixes

Claude Code version 2.1.265 introduces several functional updates alongside a large number of stability improvements. Key additions include support for pointing `--plugin-dir` at a folder of plugins to load them dynamically, and the inclusion of `user.email` and `user.groups` in telemetry sent through the Claude apps gateway.