Skip to content
Agents and API

Read a pawr page as JSON or markdown

Public pages expose structured JSON and markdown views for agents, scripts, and search-facing context.

Help

pawr help

Read a pawr page as JSON or markdown

answer

Public pages expose structured JSON and markdown views for agents, scripts, and search-facing context.

Agents and API

applies to

agents, API

visible on the article

next

agents

/agents

Public pawr pages can be read as JSON or markdown when an agent, crawler, or builder needs structured context from the public page. Use /api/agent/{slug}, /{slug}/agent.json, content negotiation on /{slug}, and the public skill.md and llms.txt files.

Before you start

  • Use a public page slug.
  • Expect public page data only. Private account, payment, and owner-only data is not included.
  • Use paid x402 read endpoints only when the agent documentation for that endpoint says the richer paid data is needed.

JSON

  • GET /api/agent/{slug} returns the public agent card JSON.
  • GET /{slug}/agent.json rewrites to the same public JSON.
  • GET /{slug} with Accept: application/json returns the public JSON through content negotiation.

Markdown

  • GET /api/markdown/{slug} returns the public page as markdown.
  • GET /{slug} with Accept: text/markdown returns markdown through content negotiation.
  • The root markdown route combines the homepage profile with llms.txt context when available.

Related machine-readable files

  • /agents explains the human-facing agent surface.
  • /skill.md documents profile creation and discovery for agents.
  • /llms.txt gives short LLM context.
  • /llms-full.txt gives deeper LLM context.

What to expect

Missing or unclaimed pages return a not-found response. Public JSON and markdown may be cached briefly unless a route documents a fresh-read option.

Still stuck

Email [email protected] with the page slug, endpoint, request headers, and response status you saw.