Skip to main content
The sequential structure is the most fundamental and commonly used execution pattern in a Pipeline.
Multiple steps are executed in order, where the output of one step (if any) can serve as the input to the next, or each step can run independently.
A standard RAG workflow can typically be constructed using only the sequential structure, allowing smooth execution from data loading to final evaluation.

​
Usage Example

https://mintcdn.com/ultrarag/T7GffHzZitf6TThi/images/yaml.svg?fit=max&auto=format&n=T7GffHzZitf6TThi&q=85&s=69b41e79144bc908039c2ee3abbb1c3bexamples/rag.yaml
# 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_init
- retriever.retriever_search
- generation.generation_init
- prompt.qa_rag_boxed
- generation.generate
- custom.output_extract_from_boxed
- evaluation.evaluate
In a sequential structure, each line in the Pipeline represents a call to a specific Tool.
The basic syntax is as follows: 
- server_name.tool_name
  • server_name: The name of the module being invoked. It must be declared in the servers section in advance.
  • tool_name: The name of the function registered within 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 structures such as loops and branches.