# API

- [Introduction](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/introduction.md)
- [Audiences](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/introduction/audiences.md): who uses what
- [Why MCP](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/introduction/why-mcp.md): the /search and /conversation removal
- [Authentification](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/authentification.md)
- [Instance API keys](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/authentification/instance-api-keys.md)
- [OAuth 2.1](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/authentification/oauth-2.1.md)
- [KAI Instance API](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/kai-instance-api.md)
- [Orchestrator](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/kai-instance-api/orchestrator.md): The Orchestrator endpoints drive document indexation: starting jobs, tracking their status, and managing re-indexation.
- [Documents](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/kai-instance-api/documents.md): The Documents endpoints expose metadata and indexation status for files in the instance.
- [Audit (raw primitives)](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/kai-instance-api/audit-raw-primitives.md)
- [Semantic graph](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/kai-instance-api/semantic-graph.md): The Semantic Graph endpoints expose the knowledge graph built from indexed documents: nodes, relationships, and exploration primitives.
- [Removed endpoints](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/kai-instance-api/removed-endpoints.md)
- [Retrieval API](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/retrieval-api.md)
- [Instances](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/retrieval-api/instances.md): List the KAI instances the authenticated user can access, with the instructions that apply to them.
- [Documents](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/retrieval-api/documents.md): Fetch document metadata and content from instances the authenticated user can access.
- [Semantic nodes](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/retrieval-api/semantic-nodes.md)
- [MCP Usage](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/retrieval-api/mcp-usage.md): The Retrieval API is available as an MCP server, exposing its 5 endpoints as tools for any MCP-compatible LLM client.
- [Audit API](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/audit-api.md)
- [Instances](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/audit-api/instances.md): Manage audit instances: list, retrieve, create, delete, update workflow steps and configuration. This page also covers user membership administration (adding/removing users, setting roles).
- [Documents & indexation](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/audit-api/documents-and-indexation.md)
- [Duplicates](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/audit-api/duplicates.md)
- [Mandatory questions](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/audit-api/mandatory-questions.md)
- [Questions & answers](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/audit-api/questions-and-answers.md)
- [Conflicts](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/audit-api/conflicts.md)
- [Missing subjects](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/audit-api/missing-subjects.md)
- [Stats](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/audit-api/stats.md): Retrieve audit statistics: daily activity, per-user metrics, and filtered aggregates across the audit instance.
- [MCP usage](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/audit-api/mcp-usage.md)
- [MCP Reference](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/mcp-reference.md): What is MCP & why we expose it
- [Connecting a client](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/mcp-reference/connecting-a-client.md): This page walks through installing the KAI MCP servers (Retrieval + Audit BETA) into the major MCP-compatible clients.
- [OAuth flow for MCP (detailed)](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/mcp-reference/oauth-flow-for-mcp-detailed.md): This page is the reference for anyone implementing a custom MCP client or debugging OAuth issues against KAI.
- [MCP tool catalog](https://k-ai.gitbook.io/knowledge-ai/api/api-presentation/mcp-reference/mcp-tool-catalog.md): A compact reference of every MCP tool KAI exposes, with links to the full endpoint documentation.


---

# Agent Instructions: Querying This Documentation

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

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://k-ai.gitbook.io/knowledge-ai/api/api-presentation.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.