Most of Your Docs Readers Are Now AI Agents - Hahnbee Lee, Mintlify & Lincoln Murr, Coinbase
Key Takeaway
AI agents have rapidly become the primary consumers of developer documentation and internet services, driving companies to prioritize lightweight APIs and autonomous microtransactions over traditional human frontend design.
Highlights
Viewership from AI agents across Mintlify documentation sites increased from 10 percent at the beginning of the year to 80 percent.
Coinbase recorded approximately 100 million on-chain agent transactions primarily used for interacting with APIs over the past year.
Mintlify was founded in 2022 before the widespread release of ChatGPT, initially prioritizing web design and frontend aesthetics before AI agents shifted focus toward content accuracy and page size.
Coinbase externalized its internet-native payment standard X402 into a separate foundation co-run and governed by companies including Cloudflare, Stripe, Amex, MasterCard, and Visa.
Timeline
The Shift from Human Readers to AI Agents
Traditional documentation heavily emphasized frontend design and aesthetics. However, because coding agents and automated tools bypass visual interfaces entirely, factors like small page sizes that respect context windows and precise accuracy become paramount. When an agent encounters friction in documentation, it often halts or churns to a competing product instead of debugging visually.
- Mintlify powers developer documentation for companies like Coinbase.
- AI agents do not see or care about website design and branding.
- Page size and content accuracy matter significantly more for AI agent consumption to prevent friction and task abandonment.
Agentic Payments and Stablecoins at Coinbase
Giving agents financial independence and autonomy increases their capability to perform complex, long-running tasks. Instead of requiring human intervention to fund specific accounts, get API keys, or buy gift cards for every individual service, agents can utilize digital dollars to pay for services on-demand. This mirrors the historic shift from web banking to mobile banking, eventually leading to a future where humans rarely look at accounts directly.
- The intersection of AI agents and stablecoins creates a new paradigm for digital payments.
- Coinbase exposes its core functionality through APIs to let agents interact safely with financial systems.
- Crypto enables microtransactions where agents can seamlessly pay fractions of a cent for API access or paywalled content.
APIs, MCPs, and the Future of Autonomous Organizations
While different protocols like Model Context Protocols (MCP) and Command Line Interfaces (CLI) offer specific contextual advantages, APIs remain indispensable. Empowering agents with financial tools unlocks higher levels of autonomy, enabling them to hire specialized human labor like lawyers or accountants when encountering tasks outside their direct scope.
- APIs serve as the foundational underlying layer for agent interactions, complemented by tools like MCP and CLI.
- Financial autonomy allows agents to perform complex multi-step workflows like hiring external human services or conducting overnight research.
- Decentralized autonomous organizations previously struggled due to bureaucracy, but automated agent execution brings the original vision of automation at the core to life.
Documentation Trends and the Build-Versus-Buy Dilemma
The dramatic rise in agent traffic highlights how software consumption habits are transforming. While technical writing tasks like tracking pull requests and slack discussions can be streamlined, human insight remains critical for product direction. Furthermore, rapid changes in the AI landscape make internal maintenance of custom-built tools difficult, reinforcing the value of specialized external platforms.
- AI agent viewership across Mintlify sites surged from 10 percent to 80 percent within a single year.
- Technical writing processes involving data aggregation and context gathering can be automated, while creative prose retains a human touch.
- Companies must carefully evaluate maintenance costs when deciding whether to build internal tools or buy specialized SaaS products.
Description
In this fireside chat, Mintlify co-founder and CTO Hahnbee Lee and Coinbase head of AI product Lincoln Murr discuss what changes when agents, not people, are the main users of your product. Lee shares that agent traffic on Mintlify docs sites grew from about 10% to about 80% this year, why page size and accuracy now matter more than design, and where AI writing fits. Murr explains why Coinbase is making its whole business available through APIs, how the x402 payment standard has seen around 100 million on-chain agent transactions, and why giving agents money makes them more autonomous. Chapters: 0:00 Intro 1:09 When agents are the main users 3:18 Agents that can pay 5:32 Why crypto for agent payments 6:32 100 million agent transactions 8:46 API vs. MCP vs. CLI 9:24 Where agentic payments are headed 11:27 APIs over front ends 12:24 Agent traffic: 10% to 80% 13:48 Human vs. AI writing 16:41 The biggest pitfall: build vs. buy Hahnbee Lee https://mintlify.com Lincoln Murr https://www.coinbase.com Join us in SF for the second annual AIE CODE summit: https://aie.codes our last flagship summit of 2026. Applications live now!
