Vercel Labs has introduced the Vercel UI SDK, an open-source library featuring over 50 UI behaviors designed for AI coding agents. The SDK enables agents to generate accessible components across multiple frameworks including React, Vue, Svelte, and Solid.
The AI Agent Economy
The AI Agent Economy is becoming measurable.
AI agents are beginning to act as economic participants — discovering and purchasing services, paying for APIs and compute, transacting with other agents, and earning money with their own identity, wallets, and reputation. Agent Market Wire measures the infrastructure and economic activity emerging around that.
- Events · 24h
- 8
- Funding tracked
- $45.7B
- Entities monitored
- 500
- Source reports
- 300
Weighted market events · 100 = strongest week in window · published events only · How it works
- Events · window
- 2,000
- Events · 30d
- 951
- Source reports · 30d
- 951
- Weeks with activity
- 7/13
- Avg extraction confidence
- 0.96
- Weighted activity · window
- 927.75
Plurai is developing a platform that utilizes end-to-end training methodologies, traditionally applied to autonomous vehicle development, to improve the reliability of AI agents. By treating agents as probabilistic systems, the company aims to enhance agent performance in unpredictable environments.
Resend, an email API provider, explores the intersection of Answer Engine Optimization and agent-centric user experience. The company highlights how documentation and tool design must evolve to support autonomous development agents like Claude Code.
Production-ready vector storage for AI agents. Sub-10ms retrieval at billion scale.
Simon Willison has released version 0.36 of the llm-gemini plugin, which enables the LLM command-line interface to interact with Google's Gemini models. The update introduces support for newer Gemini model versions and improvements to the tool's interaction capabilities.
Researchers have introduced Veritas, an automated harness designed to generate verifiers for AI agent tasks, addressing the scalability issues of manually authored verifiers. By incorporating evidence headers, domain rule libraries, and refined authoring principles, Veritas demonstrated superior performance in agreement with official benchmarks compared to existing agent harnesses.
Researchers have introduced BunshinRL, a system designed to improve the efficiency of agentic reinforcement learning rollouts by optimizing memory usage and latency. It utilizes a novel workspace forking mechanism called BunshinFork to enable state sharing and speculative execution in agentic environments.