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

# Read grounding sources

> Find which sources support the measured brand statements.

Find which sources support the measured brand statements.

**MCP tool:** `get_grounding_sources`. **Operation:** read. Uses the authorized connection; no project or content changes are made by this tool.

## Parameters

| Name | Type | Required | Default | Meaning and accepted values |
| - | - | - | - | - |
| `project_id` | string | Yes | `Required` | Project ID from list\_projects. Select the project in the intended workspace and environment. |
| `run_id` | string | Yes | `Required` | Run ID from list\_runs for this project. Use the exact returned value and check readiness before reading measurements. |
| `denominator` | string | No | `'grounding_pool'` | Use grounding\_pool for this workflow. The handler accepts per\_source but does not change the numeric calculation; do not present it as a separately calculated rate. |
| `aggregation_level` | string | No | `'domain'` | domain requests source-domain results; url requests individual-page results. Keep the same reporting unit when comparing measurements. |
| `group_by` | string\[] or null | No | `null` | Platform grouping is supported on the all-brand, unrestricted-sentiment path. Requests scoped to self, competitors, or a sentiment can drop the grouping with a warning. |
| `filters` | object\[] or null | No | `null` | Select measured answers with objects containing dimension and values. Use labels from list\_dimension\_options; inspect returned warnings for filters that were not applied. |
| `brand_subset` | string or null | No | `null` | Select statements about self (the target brand), competitors, or all brands. Omission uses all. |
| `sentiment_filter` | string or null | No | `null` | Restrict grounding evidence to positive, negative, or neutral sentiment; any leaves sentiment unrestricted. |

## Example call

Replace ID placeholders and example selections with values from your account. This is the `params` object for MCP `tools/call`.

```json theme={null}
{
  "name": "get_grounding_sources",
  "arguments": {
    "project_id": "PROJECT_ID",
    "run_id": "RUN_ID",
    "denominator": "grounding_pool",
    "aggregation_level": "domain",
    "brand_subset": "self"
  }
}
```

## Read the response

Measurement rows contain source domains or URLs and estimates for the requested brand and sentiment scope. Keep the selected scope with the result.

### Fields to use

These selected fields explain the example and the result to preserve. Field presence depends on the response; this is not a complete response schema.

| Field | Meaning and use |
| - | - |
| `data.brand_subset` | Brand scope used for the returned source rows. |
| `data.aggregation_level` | Requested source reporting unit; verify the returned source keys. |
| `data.denominator` | Accepted label. In this edition, changing it does not calculate a different numeric rate. |
| `data.rows[].domain_or_url` | Source associated with the returned grounding result. |
| `data.rows[].estimate` | Source estimate and uncertainty for the selected scope. This is not the citation-coverage endpoint’s estimate. |

## Response example

Selected fields from an observed IQRush Dev response. Lists are shortened and project/run IDs are placeholders. Omitted fields remain available in the actual response.

```json theme={null}
{
  "data_ready": true,
  "data": {
    "aggregation_level": "domain",
    "denominator": "grounding_pool",
    "brand_subset": "self",
    "rows": [
      {
        "domain_or_url": "allure.com",
        "group_by_values": {},
        "estimate": {
          "value": 1.1,
          "ci": {
            "lower": 0.3,
            "upper": 4,
            "level": 0.95
          },
          "n": 2,
          "stability": "insufficient",
          "stability_n": 2
        }
      }
    ]
  }
}
```

Read the actual returned scope, counts, and interval levels before interpreting these fields.

## Important behavior

Use grounding\_pool in this edition. Although per\_source is accepted, the reviewed handler only changes the denominator label; it does not calculate a separate per-source rate. Do not label these values as overall citation coverage. HTTP subject is also not applied by this handler.

Check tool and payload errors before using the result. [Response handling](/developers/guides/09-responses-and-metrics) explains estimates and empty results; [errors and recovery](/developers/guides/13-errors) covers invalid requests and temporary failures.

HTTP equivalent: [Read grounding sources](/developers/http/http-get-grounding-sources). HTTP parameters and defaults can differ.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.