---
title: "act: ground"
description: "exactly one of a capture, an entity, a text or an address, and a breadth (`GroundInput`) to the grounding with its hypotheses (`GroundingAnswer`), the channels that ran and the ones not ready with their reason"
section: reference
address: /reference/act/ground
contract: 0476e35e6e275db5
---

# act: ground

exactly one of a capture, an entity, a text or an address, and a breadth (`GroundInput`) to the grounding with its hypotheses (`GroundingAnswer`), the channels that ran and the ones not ready with their reason

[figure: The caller sends GroundInput to /v1/act/ground. The platform returns a card whose body is GroundingAnswer.]

<span class="op-method">POST</span><span class="op-path">/v1/act/ground</span>

Ground a subject at a breadth. The call needs the user rung under a token with the act grant, and it is served by the platform.

The route takes [GroundInput](/concepts/groundinput) and returns a card whose body is [GroundingAnswer](/concepts/groundinganswer).



## Request body
The request body is [GroundInput](/concepts/groundinput).



| field | type | required | note |
|---|---|---|---|
| `capture` | `hash` | no | the artefact's hash |
| `entity` | `id` | no | an entity of the record |
| `text` | `text` | no | a text a person or the agent pasted, cut to 4,000 characters |
| `address` | `text` | no | an address of a page the record may hold as a capture |
| `breadth` | `one of exact \| similar \| related \| broad` | no | the person's default when absent |



## Response
The route returns a card whose body is [GroundingAnswer](/concepts/groundinganswer). Every card ends with a foot that states the basis of each number, the scope of the call, and the time when the facts were true.

## Headers


| header | required | meaning |
|---|---|---|
| `Scale-Scope` | yes | the scope the call runs in: tenant/<id>, then /brand/<id>, then /strategy/<id> |
| `Scale-As-True-On` | no | the date the facts must have been true on; defaults to now |
| `Scale-As-Known-On` | no | the date the facts must have been known on; defaults to now |
| `Scale-Key` | no | for a change, the key made from the input; a repeat under the same key returns the unit held and writes nothing |



A caller needs the user rung and a token with the act grant. The platform serves this route.

## Example
The example sends the smallest body that GroundInput allows. The build validates it against the schema of GroundInput.

<details class="wire"><summary>Example</summary>





```sh
curl -X POST https://api.scaleintelligence.co/v1/act/ground \
  -H "Scale-Key: $SCALE_KEY" -H "Scale-Scope: tenant/<id>" \
  -H "content-type: application/json" \
  -d '{}'
```




```typescript
import { client } from "@scale/sdk";
const answer = await client.act.ground({});
// answer.body is a GroundingAnswer. answer.foot holds the basis, the scope and the time.
```




```python
import requests
answer = requests.post("https://api.scaleintelligence.co/v1/act/ground",
    headers={"Scale-Key": KEY, "Scale-Scope": "tenant/<id>"},
    json={}).json()
```




```rust
let answer: Card = client.post("https://api.scaleintelligence.co/v1/act/ground")
    .header("Scale-Key", key).header("Scale-Scope", "tenant/<id>")
    .json(&GroundInput { /* the fields of the page */ }).send().await?.json().await?;
```




</details>

## MCP tool
The MCP tool `si.act.ground` takes the same input and returns the same card. The tool changes data and needs a grant. An MCP host that renders cards draws this card from `ui://scale-intelligence/cards/GroundingAnswer`. The [tool's page](/reference/tools/si-act-ground) describes it.
