> 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/router.md). # router ### Schema Definition ```yaml type: object required: - type - name - inputs properties: type: type: string enum: [router] description: Block type identifier name: type: string description: Display name for this router block inputs: type: object required: - prompt - model - apiKey properties: prompt: type: string description: Instructions for routing decisions and criteria model: type: string description: AI model identifier (e.g., gpt-4o, gemini-2.5-pro, deepseek-chat) apiKey: type: string description: API key for the model provider (use {{ENV_VAR}} format) temperature: type: number minimum: 0 maximum: 2 description: Model temperature for routing decisions default: 0.3 azureEndpoint: type: string description: Azure OpenAI endpoint URL (required for Azure models) azureApiVersion: type: string description: Azure API version (required for Azure models) connections: type: object description: Multiple connection paths for different routing outcomes properties: success: type: array items: type: string description: Array of target block IDs for routing destinations ``` ### Connection Configuration Router blocks use a success array containing all possible routing destinations: ```yaml connections: success: - # Target block ID option 1 - # Target block ID option 2 - # Target block ID option 3 # Additional target block IDs as needed ``` ### Examples #### Content Type Router ```yaml content-router: type: router name: "Content Type Router" inputs: prompt: | Route this content based on its type: - If it's a question, route to question-handler - If it's a complaint, route to complaint-handler - If it's feedback, route to feedback-handler - If it's a request, route to request-handler Content: model: gpt-4o apiKey: '{{OPENAI_API_KEY}}' connections: success: - question-handler - complaint-handler - feedback-handler - request-handler ``` #### Priority Router ```yaml priority-router: type: router name: "Priority Router" inputs: prompt: | Analyze the urgency and route accordingly: - urgent-queue: High priority, needs immediate attention - standard-queue: Normal priority, standard processing - low-queue: Low priority, can be delayed Email content: Route based on urgency indicators, deadlines, and tone. model: gpt-4o temperature: 0.2 apiKey: '{{OPENAI_API_KEY}}' connections: success: - urgent-queue - standard-queue - low-queue ``` #### Department Router ```yaml department-router: type: router name: "Department Router" inputs: prompt: | Route this customer inquiry to the appropriate department: - sales-team: Sales questions, pricing, demos - support-team: Technical issues, bug reports, how-to questions - billing-team: Payment issues, subscription changes, invoices - general-team: General inquiries, feedback, other topics Customer message: Customer type: model: claude-3-5-sonnet-20241022 apiKey: '{{ANTHROPIC_API_KEY}}' connections: success: - sales-team - support-team - billing-team - general-team ``` ### Advanced Configuration #### Multiple Models Router ```yaml model-selector-router: type: router name: "Model Selection Router" inputs: prompt: | Based on the task complexity, route to the appropriate model: - simple-gpt35: Simple questions, basic tasks - advanced-gpt4: Complex analysis, detailed reasoning - specialized-claude: Creative writing, nuanced analysis Task: Complexity indicators: model: gpt-4o-mini temperature: 0.1 apiKey: '{{OPENAI_API_KEY}}' connections: success: - simple-gpt35 - advanced-gpt4 - specialized-claude ``` ### Output References Router blocks don't produce direct outputs but control workflow path: ```yaml # Router decisions affect which subsequent blocks execute # Access the routed block's outputs normally: final-step: inputs: routed-result: ``` ### Best Practices * Provide clear routing criteria in the prompt * Use specific, descriptive target block names * Include examples of content for each routing path * Use lower temperature values for consistent routing * Test with diverse input types to ensure accurate routing * Consider fallback paths for edge cases --- # 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, and the optional `goal` query parameter: ``` GET https://whitepaper.aitech.io/agentforge/yaml/yaml-schema/router.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. 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.