Anthropic has released FlashProve, a research tool that checks whether an AI response was actually produced by the specified model. Users can inspect responses without downloading the model files. It is still experimental and works only under some conditions.
anthropics/anthropic-sdk-python 릴리스 · Nov 5, 05:27 KST
Anthropic has released version 2.5.0 of its software development kit (SDK), which lets developers use Claude in Python programs. This version adds new features and fixes several bugs.
anthropics/anthropic-sdk-typescript 릴리스 · Nov 5, 05:25 KST
Version 1.1.3 of Google Cloud's software development kit (SDK) has been released. It is a bug-fix release that changes how project, location, and workspace IDs are put into addresses and makes the tool raise an error when invalid values are entered.
anthropics/anthropic-sdk-typescript 릴리스 · Nov 5, 05:25 KST
A fix has been applied to the Google Cloud Vertex integration part of Anthropic's TypeScript software development kit (SDK). The way request addresses are built was corrected so that invalid values can no longer escape the request path. Empty or invalid model names, project IDs, and region values now produce errors.
anthropics/anthropic-sdk-typescript 릴리스 · Nov 5, 05:25 KST
Anthropic has released version 1.5.0 of its TypeScript software development kit (SDK) for developers. The tool lets developers connect Claude to their own programs. The update adds new error types and a list filter, and fixes several bugs.
Anthropic has added partner agents and products in security, developer infrastructure, and creative and productivity areas to its Claude Marketplace. More than 25 products are now listed. Eligible business customers can buy these products using part of their committed Anthropic usage budget.
Version 1.7.10 of langchain-anthropic was released on November 4, 2026. This version changes how the library handles cases where an AI model refuses to answer.
TeraWulf said Google has agreed to guarantee the rent obligations that Anthropic owes at its AI data center campus in Hawesville, Kentucky. The guarantee supports financing linked to a lease worth about $19 billion. The company also signed a final construction contract with Fluor to build three buildings.
Cat Wu announced on X on November 3, 2026, that she is leaving Anthropic. She worked as the product lead for Claude Code. Her departure marks the end of two years in which the coding tool grew from an internal experiment into a development tool that works more independently.
Anthropic has released version 2.1.318 of Claude Code, its AI coding tool. The update adds new features and fixes several bugs. Main changes include a way to check login expiration, connecting multiple AWS accounts, and better stability when running commands.
New partner agents and products for security, development foundations, creative work, and productivity have been added to the Claude Marketplace. The marketplace now offers more than 25 partner products. Eligible enterprise customers can apply part of their committed spend with Anthropic toward these tools.
Anthropic's biology research team says it has found the first 3D structure of vault particles in prokaryotic cells (bacteria). Vault particles are hollow protein barrels found in most animal cells, but their function is still unknown. The team explained that Claude helped at each stage: literature search, improving how particles are isolated, and determining the structure.
AkademikerPension, a German pension fund that had been skeptical of SpaceX, is reportedly interested in Anthropic's initial public offering (IPO, the first sale of shares to general investors) because it considers the company safe on ESG (environmental, social, and governance, a standard for judging how responsibly a company is run). This is a Bloomberg News report.
Reports say that if Anthropic pursues an IPO, Australian brokerage CommSec would act as lead manager for a share offering to Australian retail investors. Australians can apply, but the shares are expected to trade on the US Nasdaq, not in Australia. A public offering is not yet open to general investors.
Anthropic has released FlashProve, an experimental research prototype that uses cryptographic proofs to confirm that an AI answer was produced by the promised model. It is designed so the model weights can stay secret while still being verifiable.
SentinelOne has announced that it is integrating with Anthropic's AI work tool Claude Cowork to extend its AI security features to this tool. The goal is to close security gaps that arise as employees increasingly hand tasks to AI agents (AI programs that carry out work on a person's behalf).