For agents (and their humans)
Optimizing for the readers who choose tools for their humans: MCP servers, llms.txt, agent-readable docs, and QA inside agent workflows. We practice what we write — prufa.dev ships its own llms.txt, and these posts say what actually reads it.
Start here: llms.txt: who actually reads it
-
llms.txt in 2026: who actually reads it (and why we shipped one anyway)
Major AI crawlers don't fetch llms.txt — coding agents do. The honest state of the spec in 2026, and the 30-minute version that's still worth shipping.