---
title: "tool si.read.groundings"
description: "one capture's, one entity's, or every grounding in a place: each grounding with its hypotheses."
section: reference
address: /reference/tools/si-read-groundings
contract: 0476e35e6e275db5
---

# tool si.read.groundings

one capture's, one entity's, or every grounding in a place: each grounding with its hypotheses.

[figure: Over MCP, si.read.groundings takes GroundingsInput and returns a card whose body is [GroundingAnswer].]

each grounding with its hypotheses. The call needs the read grant at the user rung.



| | |
|---|---|
| input | [GroundingsInput](/concepts/groundingsinput) |
| result | a card whose body is [GroundingAnswer](/concepts/groundinganswer) |
| read-only | yes |
| destructive | no |
| card address | `ui://scale-intelligence/cards/GroundingAnswer` |
| HTTP route | [`/v1/read/groundings`](/reference/read/groundings) |





| field | type | required | note |
|---|---|---|---|
| `capture` | `hash` | no |  |
| `entity` | `id` | no |  |
| `place` | `one of staged \| grounding \| waiting for a person \| committed \| discarded \| all` | no |  |