> ## Documentation Index
> Fetch the complete documentation index at: https://docs.opennous.cloud/llms.txt
> Use this file to discover all available pages before exploring further.

# Get Context

> Engineered context for one task about one entity. The ranked, claim-tagged block your agent consumes before acting.

The headline endpoint. Pass who you're working on and what you're about to do — get back a focused, ranked context block: the facts that matter, each with confidence and freshness, plus the recent timeline, buying-group stakeholders, and open predictions.

**Call this before drafting outreach, preparing for a meeting, or making any decision about a person or company.**

## Request

```bash theme={null}
curl -X POST https://api.opennous.cloud/v2/context \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{
    "focus": "sarah@acme.com",
    "intent": "follow_up",
    "budget_tokens": 1500
  }'
```

### Body

<ParamField body="focus" type="string" required>
  Who to look up. Accepts an **email**, **domain**, **LinkedIn URL**, **entity UUID**, or a **name**. A name may match several people — the response will surface candidates.
</ParamField>

<ParamField body="intent" type="string">
  What you're about to do. Shapes which context surfaces. One of:

  * `account_review` *(default)* — general lookup
  * `draft_email` — outreach prep
  * `follow_up` — re-engagement after a gap
  * `meeting_prep` — pre-meeting brief
  * `call_prep` — pre-call brief
</ParamField>

<ParamField body="budget_tokens" type="number">
  Approximate token budget for the returned context block. The pipeline ranks and compresses to fit.
</ParamField>

## Response

```json theme={null}
{
  "entity": { "id": "...", "type": "person", "name": "Sarah Chen" },
  "summary": "Sarah is evaluating for Q3; primary concern is Salesforce migration.",
  "claims": [
    {
      "property": "title",
      "value": "VP Sales",
      "confidence": 0.92,
      "freshness": "fresh",
      "epistemic_class": "observed",
      "sources": ["hubspot", "linkedin"]
    }
  ],
  "workspace": [
    { "property": "ICP", "value": "GTM Engineers or RevOps" }
  ],
  "timeline": [
    { "tier": "recent", "when": "2026-05-21T15:00:00Z", "type": "interaction.email_replied", "summary": "Interested but budget waiting on Q3." }
  ],
  "stakeholders": [
    { "name": "Maria Lopez", "role": "CFO" }
  ],
  "predictions": [
    { "kind": "icp_fit", "value": 0.84, "confidence": 0.78 }
  ],
  "meta": { "claims_returned": 12, "token_estimate": 1480 }
}
```

### Ambiguous focus

If `focus` is a name that matches several people, you'll get:

```json theme={null}
{
  "status": "ambiguous",
  "candidates": [
    { "entity_id": "...", "name": "Sarah Chen", "detail": "VP Sales @ Acme" },
    { "entity_id": "...", "name": "Sarah Chen", "detail": "Designer @ Pixel" }
  ]
}
```

Pick one and re-call with `focus: "<entity_id>"`.

## When to call it

* **Before any outbound action** on a known contact
* **At the start of meeting prep** with a saved intent for the use case
* **NOT** for "what do we know about X" questions where you want raw data — use [`/v2/accounts/:id`](/public-api/accounts) for that