# Create reindex

URL: https://docs.hanzo.ai/docs/openapi/knowledge/post-knowledge-reindex

> Rebuilds the caller org's retrieval from its documents: every passage is removed, and every page, memory and source is cut into passages and embedded again with the configured model; the lexical index is reconciled to the same documents.

`POST /v1/knowledge/reindex`

> [Knowledge →](https://docs.hanzo.ai/docs/openapi/knowledge) · [All API references →](https://docs.hanzo.ai/docs/openapi) · [Get started →](https://docs.hanzo.ai/docs/guides)

|               |                                             |
| ------------- | ------------------------------------------- |
| **Address**   | `https://api.hanzo.ai/v1/knowledge/reindex` |
| **Method**    | `POST`                                      |
| **Operation** | `post_knowledge_reindex`                    |
| **Auth**      | `Authorization: Bearer $HANZO_API_KEY`      |

Rebuilds the caller org's retrieval from its documents: every passage
is removed, and every page, memory and source is cut into passages and
embedded again with the configured model; the lexical index is reconciled to
the same documents. It is what an operator runs after the embedding model
changes — passages of another model are never compared with the query, so
until then they are unread — and what puts an org's retrieval right after an
outage of the ai plane. It requires ORG ADMIN and runs inline: an org's
knowledge is a few thousand documents, and the answer is the count.

The request has no body. Response: {"vectors": 412, "lexical": 412, "removed": 3, "failed": 0}

## Request

The document declares no body for `POST /v1/knowledge/reindex`. The handler is typed in cloud but its shape is not yet emitted, so the fields are not listed here — ask MCP's `describe` for `post_knowledge_reindex`, which answers from the running route.

## Response

| Status    | Body                 | Meaning |
| --------- | -------------------- | ------- |
| `200`     | knowledge.reindexOut | ok      |
| `default` | problem-details      | refused |

`200` body — 4 fields.

| Field     | In   | Type            | Always | Description                                                                                                                                                                                             |
| --------- | ---- | --------------- | ------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `failed`  | body | integer (int64) | —      | Failed is how many documents could not be indexed; each is logged with its name and has no passages — the rest of the rebuild went on without it.                                                       |
| `lexical` | body | integer (int64) | —      | Lexical is how many rows the org's lexical index holds now; 0 in a deployment without the index app.                                                                                                    |
| `removed` | body | integer (int64) | —      | Removed is how many rows the lexical index held for documents that no longer exist; 0 without the index app.                                                                                            |
| `vectors` | body | integer (int64) | —      | Vectors is how many documents were read back, cut into passages, embedded and written to the org's passage store, which was emptied first — so it now holds these documents' passages and nothing else. |

Failure carries the platform error shape — see [Errors](https://docs.hanzo.ai/docs/errors).

## Examples

**CLI**

```bash
hanzo knowledge reindex
```

**SDK**

**.ts**

```ts
import { Configuration, KnowledgeApi } from 'hanzoai';

const api = new KnowledgeApi(new Configuration({ accessToken: process.env.HANZO_API_KEY }));
const { data } = await api.postKnowledgeReindex();
```

**.py**

```python
from hanzoai.cloud import ApiClient, Configuration
from hanzoai.cloud.api import KnowledgeApi

client = ApiClient(Configuration(access_token=os.environ["HANZO_API_KEY"]))
result = KnowledgeApi(client).post_knowledge_reindex()
```

**.go**

```go
cfg := hanzoai.NewConfiguration()
cfg.AddDefaultHeader("Authorization", "Bearer "+os.Getenv("HANZO_API_KEY"))
client := hanzoai.NewAPIClient(cfg)

resp, _, err := client.KnowledgeAPI.PostKnowledgeReindex(context.Background()).Execute()
if err != nil {
	return err
}
```

**.rs**

```rust
use hanzo_client::apis::{configuration::Configuration, knowledge_api};

let mut cfg = Configuration::new();
cfg.bearer_access_token = std::env::var("HANZO_API_KEY").ok();

let result = knowledge_api::post_knowledge_reindex(&cfg, Default::default()).await?;
```

**.java**

```java
import ai.hanzo.cloud.ApiClient;
import ai.hanzo.cloud.api.KnowledgeApi;

ApiClient client = new ApiClient();
client.setBearerToken(System.getenv("HANZO_API_KEY"));

var result = new KnowledgeApi(client).postKnowledgeReindex();
```

**HTTP**

```bash
curl -X POST https://api.hanzo.ai/v1/knowledge/reindex \
  -H "Authorization: Bearer $HANZO_API_KEY"
```

**MCP**

MCP reaches **knowledge** through the `knowledge` tool, which names its 9 operations with its own verbs — this one among them, under a name only MCP declares. `describe` explains any of them:

```bash
curl -X POST https://api.hanzo.ai/v1/mcp \
  -H "Content-Type: application/json" \
  -d '{
       "jsonrpc": "2.0",
       "id": 1,
       "method": "tools/call",
       "params": {
         "name": "describe",
         "arguments": {
           "op": "list_knowledge_connectors"
         }
       }
     }'
```

---

[Knowledge API](https://docs.hanzo.ai/docs/openapi/knowledge) · [All Hanzo APIs](https://docs.hanzo.ai/docs/openapi) · [Interactive reference](https://docs.hanzo.ai/docs/openapi/reference)