The sequential structure is the most basic and common execution method in the UltraRAG Pipeline. Multiple steps are executed sequentially, where the output of the previous step (if any) can be used as the input for the next step, or each step can be executed independently. Building a basic RAG workflow can usually be accomplished using only the serial structure.

​
Example: Vanilla RAG Pipeline

/images/yaml.svgexamples/rag.yaml
# Vanilla RAG demo

# MCP Server
servers:
  benchmark: servers/benchmark
  retriever: servers/retriever
  prompt: servers/prompt
  generation: servers/generation
  evaluation: servers/evaluation
  custom: servers/custom

# MCP Client Pipeline
pipeline:
- benchmark.get_data
- retriever.retriever_deploy_search
- prompt.qa_rag_boxed
- generation.generate
- custom.output_extract_from_boxed
- evaluation.evaluate

​
Explanation

In the sequential structure, each line represents calling a registered Tool, with the basic syntax: server_name.tool_name
  • server_name: indicates the module name being called, which must be predefined in the servers section;
  • tool_name: indicates the function registered in that module using the @tool(…) or @prompt(…) decorator.
This structure is suitable for most single-turn question answering or reasoning tasks and serves as the foundation for understanding more complex workflows (such as loops and branches).