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

# Use local in-memory storage

The `local` backend stores knowledge in an in-memory vector index from LlamaIndex. Knowledge text is embedded by an embedding model, written to an in-process vector index, and retrieved by similarity. It requires no external database but does require an embedding service, is the **default backend** for `KnowledgeBase`, and is ideal for local debugging and quick validation.

<Warning>
  Data lives only in process memory and is lost when the program exits. Use a persistent backend in production.
</Warning>

## When to use

* Local debugging, teaching, or quickly validating a RAG flow;
* No external vector store required — only an embedding model.

## Dependencies

```bash lines theme={null}
pip install "veadk-python[extensions]"
```

## Usage

Before running, configure `MODEL_EMBEDDING_NAME`, `MODEL_EMBEDDING_DIM`, `MODEL_EMBEDDING_API_BASE`, and `MODEL_EMBEDDING_API_KEY`. The `extensions` extra includes llama-index, embedding adapters, and vector-store connectors. Text is sent to the configured embedding service. For BytePlus or another provider, explicitly set the matching endpoint, model, and credentials; the default Ark endpoint does not automatically switch

```python lines theme={null}
from veadk import Agent
from veadk.knowledgebase import KnowledgeBase

kb = KnowledgeBase(backend="local", index="company_faq")
assert kb.add_from_text("The standard annual leave is 15 days per year, available after one year of service.")
for entry in kb.search("annual leave", top_k=3):
    print(entry.content)

agent = Agent(
    name="demo",
    instruction="Answer the user; when needed, use the `load_knowledgebase` tool to search the knowledge base.",
    knowledgebase=kb,
)
```

You can also pass embedding config explicitly via `backend_config`:

```python lines theme={null}
from veadk.knowledgebase import KnowledgeBase
from veadk.configs.model_configs import EmbeddingModelConfig

kb = KnowledgeBase(
    backend="local",
    index="company_faq",
    backend_config={
        "index": "company_faq",
        "embedding_config": EmbeddingModelConfig(
            name="doubao-embedding-vision-250615",
            dim=2048,
        ),
    },
)
```

## Parameters

### KnowledgeBase parameters

| Parameter | Type | Default | Description |
| - | - | - | - |
| `backend` | `str \| BaseKnowledgebaseBackend` | `"local"` | Set to `local` or pass a backend instance |
| `backend_config` | `dict` | `{}` | Must include index when nonempty; does not merge the outer index |
| `index` | `str` | `""` | Index name; falls back to app\_name when no configuration dictionary is supplied |
| `app_name` | `str` | `""` | Fallback for index; not a user authorization filter |
| `top_k` | `int` | `10` | Default result count; search(top\_k=0) uses this value |
| `name` | `str` | `"user_knowledgebase"` | Knowledge-base name shown to the agent |
| `description` | `str` | `"This knowledgebase stores some user-related information."` | Explains the knowledge base to the agent |
| `enable_profile` | `bool` | `False` | Enables document profiles; generate profile files first, or leave disabled for ordinary retrieval |
| `query_with_user_profile` | `bool` | `False` | Uses the agent’s Viking long-term memory profile to guide queries; the knowledge backend itself need not be Viking |

### Constructor parameters

`backend_config` supports the following settings:

| Parameter | Type | Default | Description |
| :- | :- | :- | :- |
| `index` | `str` | No default; provided by `KnowledgeBase` | Knowledge base index name. The local backend imposes no naming restrictions. |
| `embedding_config` | `EmbeddingModelConfig` | Read automatically from `MODEL_EMBEDDING_*` env vars | Embedding model config. |

### Embedding config

`embedding_config` is an `EmbeddingModelConfig` with env prefix `MODEL_EMBEDDING_`:

| Field | Env var | Type | Default | Description |
| :- | :- | :- | :- | :- |
| `name` | `MODEL_EMBEDDING_NAME` | `str` | `doubao-embedding-vision-250615` | Embedding model name. |
| `dim` | `MODEL_EMBEDDING_DIM` | `int` | `2048` | Embedding vector dimension. |
| `api_base` | `MODEL_EMBEDDING_API_BASE` | `str` | `https://ark.cn-beijing.volces.com/api/v3/` | API base of the embedding service. |
| `api_key` | `MODEL_EMBEDDING_API_KEY` | `str` | Falls back to `MODEL_AGENT_API_KEY`, then an auto-fetched Ark token | Key for accessing the embedding service. |

## Environment variables

```bash lines theme={null}
# Embedding model
export MODEL_EMBEDDING_NAME="doubao-embedding-vision-250615"
export MODEL_EMBEDDING_DIM=2048
export MODEL_EMBEDDING_API_KEY="your-ark-api-key"
```

<Note>
  The local backend depends only on an embedding model and requires no database connection config. Text is automatically split into chunks by file type before embedding.
</Note>

The example search should return the annual-leave policy. For empty results, check successful ingestion, matching embedding dimensions, completed server processing, network access, and permissions. Managed ingestion may not be immediately searchable. Running the configured Agent also requires model credentials