> For the complete documentation index, see [llms.txt](https://developers.laws.africa/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://developers.laws.africa/knowledge-bases/concepts.md).

# Concepts

Core concepts for working with Laws.Africa Knowledge Bases.

Knowledge Bases retrieve legal context from maintained Laws.Africa collections. They are designed for applications that need search and grounding, not full local copies of legal content.

## Knowledge Base codes

Each Knowledge Base has a unique `code`. You use the code in API URLs:

```
POST /ai/v1/knowledge-bases/{code}/retrieve
```

List the Knowledge Bases available to your account with:

```
GET /ai/v1/knowledge-bases
```

## Places and content types

Knowledge Bases are scoped by place and content type. A place can be a country, province, municipality or other locality.

The current content types are:

* legislation;
* judgments.

For legislation, place identifiers follow the FRBR place codes used by Akoma Ntoso and the Content API. For example, `za` is South Africa and `za-cpt` is the City of Cape Town.

## Retrieve requests

A retrieve request includes:

* `text`: the query text to match;
* `top_k`: the maximum number of results to return;
* `filters`: optional metadata filters.

```json
{
  "text": "delict slip and trip",
  "top_k": 5,
  "filters": {}
}
```

`top_k` defaults to `10`. The maximum is `100`.

## Retrieve results

Each result includes:

* `content.text`: the matched legal text or summary;
* `metadata`: source information such as title, FRBR URI, dates and public URLs;
* `score`: an opaque relevance score.

Results are returned in descending relevance order. Higher scores indicate stronger matches. Treat scores as opaque non-negative ranking values; compare them only within the same response.

Use public URLs from result metadata when showing results to users or grounding AI responses. They let users inspect the source material.

## Query text

You can send a user's question directly as `text`, but AI workflows often get better matches when a model first rewrites the question as a focused legal search query.

For example, a user may ask:

```
Can my landlord lock me out without a court order?
```

Your application might query:

```
eviction lockout without court order residential tenant
```

## Plans and limits

Knowledge Bases are available on the platform's Sandbox, Build and Scale plans. Plans differ by country coverage, content availability and rate limits. See [Pricing and plans](/get-started/pricing.md) for developer-facing details.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://developers.laws.africa/knowledge-bases/concepts.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.