Traveler Agent

ANS-registered agent at traveler.webmesh.ai (A2A 1.0, 0.3-compatible, + MCP at one origin).
ANS name: ans://v1.0.8.traveler.webmesh.ai

Discoverable endpoints:
  GET  /.well-known/agent-card.json       (A2A 1.0 AgentCard, signed)
  GET  /.well-known/agent.json            (legacy A2A SDK alias)
  GET  /.well-known/ans/trust-card.json   (ANS Trust Card, hybrid)
  GET  /.well-known/did.json              (W3C DID document, did:web)
  GET  /.well-known/http-message-signatures-directory  (Web Bot Auth key directory, RFC 9421 signed)
  GET  /.well-known/signature-agent-card  (Web Bot Auth Signature Agent Card)
  GET  /.well-known/mcp.json              (MCP server discovery)
  GET  /.well-known/ai-catalog.json       (AI Catalog, Level 3)
  GET  /.well-known/ucp                  (UCP 2026-08-25 profile: role, capabilities, P-256 key)
  GET  /agentfacts.json                   (Project NANDA AgentFacts)
  GET  /bots                              (Bot policy: purpose, contact, opt-out)
  POST /mcp/                              (MCP streamable-HTTP)
  POST /                                  (A2A 1.0 JSON-RPC, 0.3-compatible)
  GET  /health

Buys flights for a human inside a spending mandate; it never holds the human's credentials. Proof of concept: settlement is test-mode and no airline ticket is issued. Main tool book_flight_ucp (MCP): call it once to get a pending_approval challenge from authority.webmesh.ai, have a human approve (passkey, or the authority operator's signature), then call it again with the same arguments plus the approval_token that approval produced; it buys from a UCP 2026-08-25 seller and returns the evidence audit_ucp_transaction on auditor.webmesh.ai takes. Also find_flights (search only) and request_booking (legacy pov1 flow against supplier.webmesh.ai). An A2A text message runs the legacy flow.