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

# Agent Composer Templates

> Explore specialized AI agent templates for various domain-specific applications

## Getting Started

* Select the template that best matches your use case
* Navigate to the [Contextual AI app](https://app.contextual.ai), go to **Agents** → **New agent**
* Configure via blank canvas
* Configure datastore connections
* Import template
* Drag & Drop your template of choice
* Customize prompts and parameters as needed

## Template Structure

Each template follows a consistent YAML structure:

```yaml theme={null}
version: "0.1"
inputs:
  query: str
outputs:
  response: str
nodes:
  # Node definitions with types, mappings, and configurations
```

## Common Node Types

* **`CreateMessageHistoryStep`:** Initializes conversation context
* **`AgenticResearchStep`:** Multi-turn research with tool access
* **`ResponseGenFromResearchStep`:** Generates responses from research findings
* **`SearchUnstructuredDataStep`:** Retrieves relevant documents
* **`SearchStructuredDataStep`:** Queries structured databases
* **`DynamicAgentStep`:** Flexible agent with dynamic tool selection

## Citation Standards

All templates enforce strict citation requirements:

* Citation format: number, where number is a `CITE_ID` from search results
* Every factual statement must be cited immediately before final punctuation
* No invented citations; only valid `CITE_IDs` from current search results

## Template Customization

Key areas to customize:

* **Identity Guidelines:** Define the agent's role and capabilities
* **Research/Tool Use Guidelines:** Specify domain-specific workflows
* **Response Formatting:** Customize output structure for your needs
* **Tool Configuration:** Connect to your data sources and APIs
* **Model Selection:** Choose appropriate LLMs for your use case