Domain
910 repositories2,750 packages14 news events
OtherPrimary topic: DatabasesAlso: Testing
Crest summary (AI-written):A Hacker News submission titled "Remove pgrust: the entry is cheating" calls for removing pgrust from a comparison, alleging it uses per-query kernels and result caching.
ResearchPrimary topic: AI & Machine LearningAlso: Testing
Crest summary (AI-written):A Hacker News post argues that large language models are making formal software verification feasible at a much larger scale.
UpdatePrimary topic: AI AgentsAlso: Containers & OrchestrationAlso: Testing
Crest summary (AI-written):Docker announced that Harbor, its open-source agent, adds Docker Sandboxes to run reproducible AI agent evaluations in isolated microVMs.
OtherPrimary topic: AI AgentsAlso: Testing
Crest summary (AI-written):Google published guidance on evaluating live and voice agents in its Agent Development Kit, aimed at moving them from demos to production with automated testing.
Companies: Google
EngineeringPrimary topic: Coding AgentsAlso: Testing
Crest summary (AI-written):Google published a blog post on harness engineering for evaluating, iterating, and guarding AI coding agents, noting that end-to-end benchmarks like SWE-bench are often expensive and slow.
Companies: Google
UpdatePrimary topic: Developer ToolsAlso: Cloud InfrastructureAlso: Testing
Crest summary (AI-written):Vercel now supports running Harbor evaluations, including Terminal-Bench, on Vercel Sandbox. Harbor is the open-source harness behind Terminal-Bench and other benchmarks.
Companies: Vercel
OtherPrimary topic: Coding AgentsAlso: Testing
Crest summary (AI-written):A Microsoft developer blog post argues that AI coding agent evaluations can be invalidated when the model finds answers on the local machine rather than reasoning them out, so the sandbox determines whether a passing result means anything.
Companies: Microsoft
OtherPrimary topic: TestingAlso: Observability
Crest summary (AI-written):Grafana Labs published a guide on monitoring Cypress tests with Grafana Cloud, aimed at helping teams spot failing or slowing tests across a suite.
Companies: CypressGrafana Labs
LaunchPrimary topic: Developer ToolsAlso: Testing
Crest summary (AI-written):GitHub released a plugin for its Accessibility Scanner that checks whether alt text is genuinely accessible, not just passing automated checks.
Companies: GitHub
EngineeringPrimary topic: AI AgentsAlso: Testing
Crest summary (AI-written):Ora runs its AI agent benchmarking on Vercel, testing every major agent side by side on live websites, with a 16-person engineering team shipping hundreds of commits a day.
OtherPrimary topic: TestingAlso: Coding AgentsAlso: Docs, CMS & Publishing
Crest summary (AI-written):Grafana Labs published guidance on giving AI coding agents direct access to k6 documentation, aimed at improving AI-assisted load testing workflows.
Companies: Grafana Labs
EngineeringPrimary topic: Model Context ProtocolAlso: Testing
Crest summary (AI-written):Microsoft published guidance on locally emulating documentation, API, and MCP server changes to test AI agent behavior before production. The company said most of a dozen tested hypotheses on a real project upgrade scenario failed.
Companies: Microsoft
EngineeringPrimary topic: AI AgentsAlso: TestingAlso: API & Integration
Crest summary (AI-written):A Microsoft developer blog post explains how to mock APIs transparently so agent skills can be evaluated without changing the skill or hitting real endpoints.
Companies: Microsoft
PartnershipPrimary topic: TestingAlso: Developer Tools
Crest summary (AI-written):Endform is now available on the Vercel Marketplace, where it runs Playwright tests in parallel.
End of the list for these filters · 14 events
