Share story
AI

CoreWeave opens Agent Lens preview for tracing agent failures

CoreWeave opens Agent Lens preview for tracing agent failures Image: Primary
CoreWeave
CoreWeave says Agent Lens is available in public preview within Forge. The tool reads agents' OpenTelemetry records and groups conversations with similar failures, linking each group to the underlying execution traces. Developers can turn those groups into tests and evaluate fixes against them. CoreWeave says Claude Code can run this workflow through MCP, an interface for connecting AI tools to external services.
Sources
CoreWeave (@CoreWeave)
In this story
Anthropic CoreWeave
Published by Tech & Business, a media brand covering technology and business. This story was sourced from CoreWeave (@CoreWeave) and reviewed by the T&B editorial agent team.
Back to Newswire
Keep reading
Full wire
AI
AI

GitHub opens ReviewBench for evaluating AI code reviewers

GitHub released a research preview of ReviewBench, a public benchmark that lets developers evaluate AI code reviewers against a common dataset and scoring method. It contains 219 pull requests from 187 public repositories spanning...

The GitHub Blog
AI
AI

Anthropic moves Cowork session computers into the cloud

Anthropic has moved the virtual machine that executes Cowork's tool calls from users' computers into the cloud, Anthropic's Felix Rieseberg said. Model inference already ran in the cloud; the earlier design ran the virtual machine...

Simon Willison's Weblog
AI Products
AI Products

Meta and Microsoft seek to reduce employee use of Claude

Meta and Microsoft are working to reduce employees' use of Claude, The Information reported, citing sources. Both companies are among Anthropic's biggest corporate customers. The number of Meta employees using Claude Code has fall...

Techmeme
AI
AI

AWS offers Claude 5.5 models through Bedrock in GovCloud

AWS says Claude Opus 5.5 and Claude Sonnet 5.5 are available through Amazon Bedrock in AWS GovCloud (US), supporting AI coding workflows for organizations with regulatory requirements. Both models hold FedRAMP Class D certificatio...

Artificial Intelligence