> ## Documentation Index
> Fetch the complete documentation index at: https://docs.intellibase.dev/llms.txt
> Use this file to discover all available pages before exploring further.
# Quickstart
> Get started with Intellibase in 5 minutes
This guide will walk you through creating your first Intellibase project, ingesting data, and querying it.
## Step 1: Get Your API Key
To use Intellibase, you need an API key:
1. Log in to the [Intellibase Dashboard](https://app.intellibase.dev)
2. Navigate to **API Keys**
3. Click **Create API Key**
4. Copy your key (it starts with `ib-`)
Store your API key securely. It provides full access to your projects and data.
Pre-configured API requests ready to use with your API key
## Step 2: Create a new Project
Intellibase offers two project modes:
**Best for**: Capturing complex multi-hop relationships, hierarchal understanding, temporal reasoning and high accuracy.
**Optionally define your custom ontology** - You can define a custom ontology for your knowledge graph (with help from AI) so that you get
a highly dependable graph that captures all the information that is important to your agent without any unnecessary noise.
1. Use the [suggest ontology endpoint](/api-reference/ontologies/suggest) to get a suggested ontology by describing the use case or agent persona, optionally
sample expected input/ ouput that the agent will be expected to perform.
```bash theme={null}
# Let AI suggest an ontology based on your use case
curl -X POST https://api.intellibase.dev/api/v1/ontologies/suggest \
-H "Authorization: Bearer ib-your-api-key" \
-H "Content-Type: application/json" \
-d '{
"usecase": "Track software engineering projects, developers, and features"
}'
```
2. Use the [create ontology endpoint](/api-reference/ontologies/create) to save the ontology into Intellibase and generate an ontology ID.
```bash cURL theme={null}
curl -X POST https://api.intellibase.dev/api/v1/ontologies \
-H "Authorization: Bearer ib-your-api-key" \
-H "Content-Type: application/json" \
-d '{
"name": "Software Engineering",
"schema": {
"ontology": {
"hierarchy": [
{
"level": 0,
"node_types": {
"Developer": {
"properties": ["name", "role", "team"],
"description": "Software developer"
},
"Feature": {
"properties": ["name", "status", "deadline"],
"description": "Software feature"
}
}
}
],
"edge_types": [
{
"name": "WORKS_ON",
"source_types": ["Developer"],
"target_types": ["Feature"],
"properties": ["role"],
"temporal": true,
"description": "Developer works on feature"
}
]
}
}
}'
```
3. Then create your project with the ontology ID you got from the previous step:
```bash theme={null}
curl -X POST https://api.intellibase.dev/api/v1/projects \
-H "Authorization: Bearer ib-your-api-key" \
-H "Content-Type: application/json" \
-d '{
"name": "Engineering Knowledge Base",
"ontology_id": "ba879....",
}'
```
Create a project by providing a name. If you don't provide an ontology ID during project creation,
the knowledge graph generated will not have a fixed/ static ontology and will be generated based on
the data being ingested into the engine.
```bash theme={null}
curl -X POST https://api.intellibase.dev/api/v1/projects \
-H "Authorization: Bearer ib-your-api-key" \
-H "Content-Type: application/json" \
-d '{
"name": "Engineering Knowledge Base",
"ontology_id": "ba879....",
}'
```
**Best for**: Traditional RAG, semantic search, getting started quickly
**No ontology required** - just ingest and query!
```bash theme={null}
curl -X POST https://api.intellibase.dev/api/v1/projects \
-H "Authorization: Bearer ib-your-api-key" \
-H "Content-Type: application/json" \
-d '{
"name": "My First Project",
"vector_only": true
}'
```
Save the `project_id` from the response - you'll need it for all subsequent operations.
## Step 3: Ingest data into your project
Now let's add some data to your project. We currently only support text input, we'll soon be adding support for more input data types.
```bash theme={null}
curl -X POST https://api.intellibase.dev/api/v1/projects/{project_id}/ingest \
-H "Authorization: Bearer ib-your-api-key" \
-H "Content-Type: application/json" \
-d '{
"text": "Alice joined the engineering team in January 2024. She is working on the authentication feature, which is part of the security module. The feature is scheduled to launch in March 2024.",
"source_doc_id": "doc-001",
}'
```
The API returns a job ID:
```json theme={null}
{
"job_id": "job-abc123",
"status": "pending",
"created_at": "2024-01-15T10:30:00Z"
}
```
## Step 4: Check Ingestion Status
Monitor your ingestion job:
```bash theme={null}
curl https://api.intellibase.dev/api/v1/projects/{project_id}/jobs/{job_id} \
-H "Authorization: Bearer ib-your-api-key"
```
Response:
```json theme={null}
{
"job_id": "job-abc123",
"project_id": "proj-xyz",
"status": "completed",
"chunks_processed": 3,
"nodes_created": 5,
"edges_created": 4,
"started_at": "2024-01-15T10:30:00Z",
"completed_at": "2024-01-15T10:30:15Z"
}
```
When `status` is `completed`, your data is ready to query!
## Step 5: Query Your Data
Ask natural language questions to fetch the most relevant nodes, edges and chunks.
The static query mode is designed for low latency and does not use AI during retrieval.
The results are still of very high quality and better than a basic vanilla vector RAG system.
```bash theme={null}
curl -X POST https://api.intellibase.dev/api/v1/projects/{project_id}/query \
-H "Authorization: Bearer ib-your-api-key" \
-H "Content-Type: application/json" \
-d '{
"query": "What features is Alice working on?",
"mode": "static"
}'
```
The dynamic query mode is designed for high accuracy with a slight trade-off in terms of latency.
It uses AI to retrive the most relevant information from the context engine.
```bash theme={null}
curl -X POST https://api.intellibase.dev/api/v1/projects/{project_id}/query \
-H "Authorization: Bearer ib-your-api-key" \
-H "Content-Type: application/json" \
-d '{
"query": "What features is Alice working on?",
"mode": "dynamic"
}'
```
Response:
```json theme={null}
{
"query": "What features is Alice working on?",
"strategy_used": "hybrid",
"results": {
"summary": "Alice is working on the authentication feature, which is part of the security module and scheduled to launch in March 2024.",
"entities": [
{
"type": "Person",
"name": "Alice",
"properties": {
"role": "Engineer",
"joined": "January 2024"
}
},
{
"type": "Feature",
"name": "Authentication Feature",
"properties": {
"module": "Security",
"launch_date": "March 2024"
}
}
],
"relationships": [
{
"source": "Alice",
"target": "Authentication Feature",
"type": "WORKS_ON"
}
],
"chunks": [
{
"text": "She is working on the authentication feature...",
"similarity": 0.89
}
]
}
}
```