INSIGHTS & UPDATES
Perspectives on AI agents, WebMCP, and building the next generation of agent-ready web applications.
How to build observability for AI agents inside your B2B SaaS product. Covers the SubmitEvent.agentInvoked flag, tool call logging, session attribution, permission auditing, and the metrics that matter when third-party agents use your application.
A step-by-step WebMCP implementation guide for web developers. Covers environment setup, feature detection, the @mcp-b/global polyfill, declarative and imperative tool registration, testing with the Inspector extension, and production deployment checklist.
How to design WebMCP tools that AI agents call correctly every time. Covers naming conventions, JSON Schema best practices, readOnlyHint, error handling, and the tool contract patterns that reduce hallucinations.
WebMCP and Anthropic's Model Context Protocol (MCP) are both agent integration standards — but they solve different problems. This guide explains the technical differences, when each is the right choice, and the architecture that combines both for full agent coverage.
B2B software is shifting from passive tools to active participants. Here is how AI agents are transforming product experiences and what it means for SaaS teams.
The agentic web — where AI agents use software on users' behalf — is arriving faster than most product teams realize. This guide explains what the shift means for B2B SaaS roadmaps, which capabilities to prioritize, and how to think about agent-readiness as a product strategy.
WebMCP is the open protocol that allows AI agents to read, understand, and interact with web applications. Learn how this standard works and why it matters for the future of agentic software.