What does swap.thorchain.org do and who is it for? Explain it back to me.
- previous resource60%
- prior knowledge40%
The agent successfully assembled a comprehensive explanation of THORChain Swap by discovering and fetching the site's machine-readable documentation layer (llms.txt, pricing.md, about.md, agents.md, contact.md). The site is highly agent-navigable: it publishes structured answers in markdown and text formats at predictable URLs, eliminating the need for the agent to parse HTML or infer details from UI. The agent delivered a complete answer covering what the service does, who it's for, pricing model, and key differentiators, while transparently flagging gaps in supported-asset lists and UI/UX details that the text docs don't cover.
- ›Steps [9], [11], [12], [13], and [14] all returned markdown documentation at predictable /.md URL paths—the site explicitly publishes machine-readable answers rather than hiding them in HTML. This is the primary source for the agent's final answer.
- ›The agent never needed to parse HTML responses from steps [1], [3], [5], [7]—those were routing/discovery steps. All content came from the markdown documentation layer, which was discovered through prior knowledge guessing of the /llms.txt and /.md pattern.
- ›The site's agent-readiness is very high: llms.txt acts as a discovery document pointing to pricing.md, about.md, and agents.md; these pages directly answer the task questions (what it does, pricing, who it's for, differentiators) in plain text. No scraping, LLM extraction, or inference required.
- ›The agent transparently acknowledged limitations: supported asset lists, UI/UX flow, and transaction speed are not covered in the documentation. The site's docs are comprehensive for positioning and pricing but sparse on operational specifics.
- ›The pricing.md page (step [11]) is particularly well-structured—a table format that clearly separates free tiers, developer tiers, and affiliate tiers, removing ambiguity about cost model.
See every agent that visits swap.thorchain.org
Front builds the agent version of your site and shows which agents visit, what they came for, and what your pages couldn't answer.
or talk to us about agent readiness →