> For the complete documentation index, see [llms.txt](https://whitepaper.aitech.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://whitepaper.aitech.io/agentforge/yaml/yaml-schema.md).

# YAML Schema

This section contains the complete YAML schema definitions for all available block types in Agent Forge. Each block type has specific configuration requirements and output formats.

### Core Blocks

These are the essential building blocks for creating workflows:

<table data-view="cards"><thead><tr><th>Title</th><th data-card-target data-type="content-ref">Target</th></tr></thead><tbody><tr><td>Starter Block<br>Workflow entry point supporting manual triggers, webhooks, and schedules</td><td><a href="file:///yaml/blocks/starter">file:///yaml/blocks/starter</a></td></tr><tr><td>Agent Block<br>AI-powered processing with LLM integration and tool support</td><td><a href="file:///yaml/blocks/agent">file:///yaml/blocks/agent</a></td></tr><tr><td>Function Block<br>Custom JavaScript/TypeScript code execution environment</td><td><a href="file:///yaml/blocks/function">file:///yaml/blocks/function</a></td></tr><tr><td>Response Block<br>Format and return final workflow results</td><td><a href="file:///yaml/blocks/response">file:///yaml/blocks/response</a></td></tr></tbody></table>

### Logic & Control Flow

Blocks for implementing conditional logic and control flow:

<table data-view="cards"><thead><tr><th>Title</th><th data-card-target data-type="content-ref">Target</th></tr></thead><tbody><tr><td>Condition Block<br>Conditional branching based on boolean expressions</td><td><a href="file:///yaml/blocks/condition">file:///yaml/blocks/condition</a></td></tr><tr><td>Router Block<br>AI-powered intelligent routing to multiple paths</td><td><a href="file:///yaml/blocks/router">file:///yaml/blocks/router</a></td></tr><tr><td>Loop Block<br>Iterative processing with for and forEach loops</td><td><a href="file:///yaml/blocks/loop">file:///yaml/blocks/loop</a></td></tr><tr><td>Parallel Block<br>Concurrent execution across multiple instances</td><td><a href="file:///yaml/blocks/parallel">file:///yaml/blocks/parallel</a></td></tr></tbody></table>

### Integration Blocks

Blocks for connecting to external services and systems:

<table data-view="cards"><thead><tr><th>Title</th><th data-card-target data-type="content-ref">Target</th></tr></thead><tbody><tr><td>API Block<br>HTTP requests to external REST APIs</td><td><a href="file:///yaml/blocks/api">file:///yaml/blocks/api</a></td></tr><tr><td>Webhook Block<br>Webhook triggers for external integrations</td><td><a href="file:///yaml/blocks/webhook">file:///yaml/blocks/webhook</a></td></tr></tbody></table>

### Advanced Blocks

Specialized blocks for complex workflow patterns:

<table data-view="cards"><thead><tr><th>Title</th><th data-card-target data-type="content-ref">Target</th></tr></thead><tbody><tr><td>Evaluator Block<br>Validate outputs against defined criteria and metrics</td><td><a href="file:///yaml/blocks/evaluator">file:///yaml/blocks/evaluator</a></td></tr><tr><td>Workflow Block<br>Execute other workflows as reusable components</td><td><a href="file:///yaml/blocks/workflow">file:///yaml/blocks/workflow</a></td></tr></tbody></table>

### Common Schema Elements

All blocks share these common elements:

#### Basic Structure

```yaml
block-id:
  type: <block-type>
  name: <display-name>
  inputs:
    # Block-specific configuration
  connections:
    # Connection definitions
```

#### Connection Types

* **success**: Target block for successful execution
* **error**: Target block for error handling (optional)
* **conditions**: Multiple paths for conditional blocks

#### Environment Variables

Use double curly braces for environment variables:

```yaml
inputs:
  apiKey: '{{API_KEY_NAME}}'
  endpoint: '{{SERVICE_ENDPOINT}}'
```

#### Block References

Reference other block outputs using the block name in lowercase:

```yaml
inputs:
  userPrompt: <blockname.content>
  data: <functionblock.output>
  originalInput: <start.input>
```

### Validation Rules

All YAML blocks are validated against their schemas:

1. **Required fields**: Must be present
2. **Type validation**: Values must match expected types
3. **Enum validation**: String values must be from allowed lists
4. **Range validation**: Numbers must be within specified ranges
5. **Pattern validation**: Strings must match regex patterns (where applicable)

### Quick Reference

#### Block Types and Properties

| Block Type | Primary Output  | Common Use Cases                      |
| ---------- | --------------- | ------------------------------------- |
| starter    | `.input`        | Workflow entry point                  |
| agent      | `.content`      | AI processing, text generation        |
| function   | `.output`       | Data transformation, calculations     |
| api        | `.output`       | External service integration          |
| condition  | N/A (branching) | Conditional logic                     |
| router     | N/A (branching) | Intelligent routing                   |
| response   | N/A (terminal)  | Final output formatting               |
| loop       | `.results`      | Iterative processing                  |
| parallel   | `.results`      | Concurrent processing                 |
| webhook    | `.payload`      | External triggers                     |
| evaluator  | `.score`        | Output validation, quality assessment |
| workflow   | `.output`       | Sub-workflow execution, modularity    |

#### Required vs Optional

* **Always required**: `type`, `name`
* **Usually required**: `inputs`, `connections`
* **Context dependent**: Specific input fields vary by block type
* **Always optional**: `error` connections, UI-specific fields


---

# 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 dynamically by asking a question.

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

```
GET https://whitepaper.aitech.io/agentforge/yaml/yaml-schema.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.