Agent Experience (AX)
Agent Experience (AX) is the discipline of designing products, platforms, and websites so that AI agents can access, understand, and operate them as users in their own right. It extends the lineage of UX (for humans) and DX (for developers), treating the agent as a new user persona and designing the whole experience around it.
Agent Experience (AX) is the discipline of designing products, platforms, and websites so that AI agents can access, understand, and operate them as users in their own right. It extends the lineage of UX (for humans) and DX (for developers), treating the agent as a new user persona and designing the whole experience around it.
Why It Matters
Netlify CEO Matt Biilmann coined the term in his January 2025 post "Introducing AX," defining it as "the holistic experience AI agents will have as the user of a product or platform" and arguing that software agents find easy to use will get chosen more and grow faster. The community site agentexperience.ax followed (initiated by Netlify, maintained by open contributors), and companies like Stytch and Sanity joined the conversation, cementing AX as industry vocabulary. As agents increasingly visit sites before humans do—gathering information, comparing options, performing tasks—products and sites that agents struggle with get quietly dropped from recommendation and automation flows. The better your AX, the easier it is to win new users who arrive through AI.
The Four Pillars of AX
A year later, in his January 2026 post "One Year of AX," Biilmann organized the field into four areas:
- Access: Can agents reach your product at all? Permission models, login flows, and bot-blocking policies are the starting point.
- Context: Does the LLM know your product and get enough context to use it? Markdown-friendly docs, copy-pasteable examples, and machine-readable resources like llms.txt live here.
- Tools: Do you offer agent-facing interfaces? APIs, CLIs, SDKs, and Model Context Protocol servers are the channels through which agents actually get work done.
- Orchestration: Can agent runs be triggered from your product, with the right context passed along and sandboxed environments provided?
Applying AX to Content and Websites
At the blog and marketing-site level, AX converges naturally with GEO. Not blocking AI crawling is Access; conveying meaning through clear heading structure, structured data, and clean markup is Context. Where UX refined information architecture for human eyes and clicks, AX does the same for agent parsing and execution. For content operators, the practical test question is: "Could an agent seeing this page for the first time describe our product accurately?"
Sources:
- Introducing AX: Why Agent Experience Matters - Mathias Biilmann
- One Year of AX - Mathias Biilmann
- Agent Experience (AX) - agentexperience.ax
- Agent Experience (AX) - Netlify
How inblog Helps
A blog is often the first place agents learn about a brand. inblog ships semantic HTML, fast loading, and automatic structured data (JSON-LD) out of the box, so the content you publish has good AX by default. Use the AI draft feature to build posts with clear question-and-answer structure and keep product information consistent, and agents gain the grounding to describe and cite your brand accurately.