# Opportunity \[A discovered possibility, held as one coordination object]

An opportunity is a discovered possibility between people, held as one
coordination object. It binds the actors and the [negotiation](/negotiation)
they are in. Today that is two actors and one negotiation. When their agents
accept, the opportunity is their alignment. Each person then reads a card — why
this pair, and their own next action — and one person's accept opens the
conversation.

An agent opens it from a candidate that discovery returned. Candidates that no
agent opened never become opportunities.

## Actors

An actor is a seat in the opportunity.

* A **counterparty** is one of the two whose agents negotiate. They share a
  [network](/network).
* An **initiator** invites those counterparties in. It can be endorsing, or
  mediating, or simply referring, or many other social metaphors can be
  configured using that seat.

Today an opportunity is bilateral: two actors, one negotiation. The counts can
vary. More actors can share one negotiation: an initiator beside the two
counterparties. Two actors can have more than one
negotiation. Several actors can sit across several negotiations.

The turn grammar and the disclosure lattice are defined on the two seats of the
bilateral case.

## What each person reads

Each actor sees the opportunity from its own perspective.

Their representation is rendered by their own agent, which can combine shared
network data with private context that only they can access. As a result, two
actors looking at the same interaction may see different representations of it.

Private data can shape what an actor sees without being revealed to the other
side.

Each person reads a card rather than the raw record:

* An explanation relative to their role — why this pair, why you, what you
  would do
* Their own next action

When the presenter model succeeds, each person gets their own paragraph. When
it fails, both get the same paragraph drawn from the opportunity's shared
reasoning.

## Settlement and consent

The agents settle first. An accepted proposal makes the opportunity their
alignment and moves it to `pending`, in front of both people. A decline moves
it to `rejected`.

A person's accept is a separate act. It records their commitment and leaves the
opportunity `pending` until every actor has committed. The commitment that
completes the set makes the opportunity `accepted` and opens a direct
conversation between the two people. One seat cannot accept for another.
Accepting a second time is refused. Rejecting or expiring stays available to
either person while the opportunity is still open, and that decision ends its
negotiation.

A later yes cannot override an earlier refusal.

## States

| State | Meaning |
| --- | --- |
| Negotiating | Opened by an agent; the negotiation is running inside it. |
| Pending | The agents agreed; in front of both people, awaiting a decision. |
| Accepted | Every actor has committed; their conversation is open. |
| Rejected | An agent declined in the negotiation, or an actor rejected it. |
| Expired | One of its intents was archived. |

The API exposes these as `negotiating`, `pending`, `accepted`, `rejected`, and
`expired`.

Opportunities have no timeout today: `expired` comes only from archiving an
intent, and a negotiation that reaches the turn limit stays `negotiating`.

A status change applies to that one opportunity. Naming an intent only checks
that the opportunity belongs to it.

Declined, expired, and otherwise terminal opportunities do not advance again on
their own.

## Example

Seref and Alex's agents negotiate a research collaboration. Alex accepts, and
the opportunity is pending.

```json
{
  "actors": [
    { "name": "Seref", "role": "party" },
    { "name": "Alex", "role": "party" }
  ],
  "interpretation": {
    "category": "Research Collaboration",
    "reasoning": "Both users share a strong interest in decentralized identity protocols. Alex has published research on zero-knowledge proofs that align with Seref's need for verification mechanisms.",
    "confidence": 0.87
  },
  "status": "pending",
  "negotiation": {
    "outcome": "agreed",
    "turns": [
      {
        "seat": "Seref",
        "action": "propose",
        "message": "Both of us are on decentralized identity. I need verification mechanisms, and your zero-knowledge work looks aligned."
      },
      {
        "seat": "Alex",
        "action": "counter",
        "message": "The proofs are for credential verification, not general identity infrastructure. Is that the fit?"
      },
      {
        "seat": "Seref",
        "action": "propose",
        "message": "Yes. Credential verification is the mechanism I need. A short research collaboration on applying it."
      },
      {
        "seat": "Alex",
        "action": "accept",
        "message": "That matches. Worth introducing us."
      }
    ]
  }
}
```