State of agent skills
Vercel reports that its skills.sh registry reached one million agent-skill listings and nearly 280 million installs in seven months.
source-linked1
Raw Signals
Dated, source-linked observations imported from the QWG AI archive.
These are evidence records, not finished editorial conclusions.
Key takeaway
Across 1332 observations, most signals relate to model behavior, evaluation discipline, and real-world agent operations. Strong evidence clusters around tooling, memory, routing, and productized agent execution.
agentscoding agentsagent harnessagent runtimemodelsagent protocols
What this means
Teams are moving from experiments to systems. The emphasis is on verifiable behavior, operational safety, and measurable outcomes—especially in agent memory, evaluation rigor, and forward-deployed engineering.
Filters (current page)
1-10 of 10 matching raw signals on this page
-skills-cdcc36 notableresearch
source-linked1pod-and-10afee notableresearchMulti-Region training with Amazon SageMaker HyperPod and Qumulo
AWS has published an architecture and validation report for training with Amazon SageMaker HyperPod compute in one AWS Region while keeping the dataset in another through Cloud Native Qumulo.
source-linked1e-video-53d257 notableresearchA kernel-centric path to real-time video generation on Trainium
Amazon Science reports that AWS and Reactor used the Neuron Kernel Interface to address real-time autoregressive-diffusion execution on AWS Trainium.
source-linked1e-cloud-1e7f61 notablereleaseCompass is coming to the cloud
Cohere has put Compass Cloud into private beta as a managed version of its enterprise search and retrieval platform.
source-linked1s-sites-eea943 notablereleaseMisconfiguring Turnstile by skipping backend validation leaves sites exposed to bots.
Cloudflare has launched Turnstile Spin, an agent-mediated setup flow for its Turnstile bot-protection product.
source-linked1-on-web-a50c9a notableresearchSearch Capability Leaderboard: 24 models ranked on web search
Parallel has launched the Search Capability Leaderboard, a public comparison of 24 open- and closed-source models used with web search.
source-linked1t-layer-a5f303 notableannouncementMerge picks Parallel Search for Universal Context Layer
Merge announced that its Universal Context Layer for Merge for Workforce is available and that Parallel Search is its default web-search provider.
source-linked1ter-0-2-ff2f1e notablereleasecommit-rewriter 0.2 adds support for operating on Git branches other than the repository’s default branch
commit-rewriter 0.2 adds support for operating on Git branches other than the repository’s default branch.
source-linked1aker-ai-68f039 notableresearchSpeaker-labeled transcription with WhisperX on SageMaker AI
AWS published a deployment pattern for speaker-labeled transcription on SageMaker AI.
source-linked1entcore-721d9e notableresearchBuild a multi-account AI agent with AgentCore Gateway and MCP
AWS has published a reference architecture for a multi-account AI agent.
source-linked1
No raw signals match these filters.
Source & exports
Primary indexes and programmatic access to this dataset.
Archive contractEnglish normalized copy, Telegram provenance, public sources only