Mintlify argues self-updating docs will become an 'infinite market'

Why Mintlify calls it an infinite market

Foster borrows the term "infinite market" from Alexandr Wang, who used it in an interview with YC last year to describe markets that can keep growing indefinitely. Wang's example was AWS: demand for compute keeps rising, and AWS keeps serving it. Foster says docs are on the same path and points to three trends.

The first is that companies' product surface areas are growing exponentially. According to Foster, coding agents have dramatically increased engineering throughput, which lets companies build larger products and take on more ambitious problems. He cites OpenAI, Anthropic and Notion, which he says are all building PowerPoint competitors even though presentations are not a core competency for any of them.

Foster says GitHub PR counts correlate fairly strongly with growth in a company's product surface area, and that the number of PRs on GitHub has doubled in the past year. At this rate, he writes, monthly code contributions in 2035 will be roughly 500x what they are in 2026.

Agents are the readers

The second trend is a shift in who reads documentation. Foster states that more than 75% of docs traffic now comes from agents. He describes his own work as often proxied through an agent, which builds presentations, writes docs, sends Slack messages and reviews code on GitHub.

To do that, he argues, agents need to know how to work with a product's entire surface area. Because they can process large amounts of unstructured information, they need a product explained differently than humans do. In his words, a human skims a quickstart and gives up on a 200-page API reference, while an agent can read every endpoint, parameter and edge case in seconds. His conclusion is that products should maximize the information agents can access, so agents can use them fully and correctly.

He also describes the scale of the current workload at Mintlify: its roughly 20 person product org has merged over 100 docs changes in the last two weeks. The documentation burden, he says, is already too large to manage by hand.

The primary source problem

The third trend concerns access. Foster says internal agents work best when they can read primary sources directly. His own Claude Code has access to PostHog, Datadog, customer support threads and the entire codebase, and he expects this access to matter more as models and harnesses improve.

Customers' agents cannot be given the same access. Foster notes that reading the codebase is a security risk and that Slack threads expose personal information. External agents therefore need a curated subset of internal information.

What self-updating docs are meant to do

Foster frames the resulting problem this way: companies have an enormous, exponentially growing product surface, agents are their users, and they need to share as much information as possible without exposing anything private. He expects the problem to grow as models improve and agents take on more work.

The vision for self-updating docs, he writes, is to be the "truth engine" that builds the external-facing representation of a product. He calls this a hard problem. It requires understanding most of the business to decide what should and should not be shared publicly, knowing who is who in the organization to verify information and capture knowledge shared offline, and organizing everything in the structure most accessible to the agents reading it.

The post sits alongside recent Mintlify announcements. On October 20, 2026, the company introduced Autopilot, which it describes as an always-on agent that reads your code and the context around it, drafts doc updates when something changes, and opens a pull request for review. On October 23, 2026, it added support for uploading custom skills to the agent in the web editor.

What it means

The post is an opinion piece from a vendor selling in this space, and its figures are Foster's own claims rather than independently verified data. Its core argument is that as product surfaces expand and agents become the main readers of documentation, demand will grow for tools that distill internal knowledge into an external, agent-ready form without leaking private information. For software companies, the message is that docs maintenance is becoming a problem of scale that manual processes are unlikely to meet.