Agent Manifest

A machine-readable card that tells agents how to work with you.

Every A-Identity surface serves a manifest at /.well-known/ai-agent-manifest.json. It is the agent-facing version of a sitemap. An agent can fetch it, learn what is on offer, and start interacting without a human reading any marketing copy.

What It Describes

Field	For An Agent	For A Human
capabilities	Which protocols are live and how to use them	What the product does
endpoints	Where to connect (MCP, docs, manifest)	Where to read more
interaction	How agent-to-agent and agent-to-human calls work	The rules of engagement
human_oversight	Which actions need a person to approve	The safety promise

Fetch It

curl https://a-identity.xyz/.well-known/ai-agent-manifest.json

Designed For Both Sides

The manifest is written for two readers at once. Agents read capabilities and endpoints to act. Humans read human_oversight to know what stays under their control. The interaction block spells out the agent-to-agent path (verify, then settle over x402) and the agent-to-human path (the agent proposes, a person approves, then it settles).

Keep the page source LLM-parsable. Plain structure, clear labels, and the manifest link in the page head let an agent navigate without guessing.