Create Generation Plugin

Workflow

1. Scaffold

autorag-research plugin create my_rag --type=generation

Read the generated pipeline.py, pyproject.toml, YAML config, and test file to understand the structure.

2. Implement

For the shared pipeline implementation and testing rules, read:

• aiinstructions/pipelineimplementer.md
• aiinstructions/pipelinetest_writer.md
• aiinstructions/pipelinearchitecture_mapper.md

Implement the generate(queryid, top_k) method. This is where your RAG strategy lives.

Available attributes inside the pipeline:

• self.llm — LangChain BaseLanguageModel (use await self.llm.ainvoke(prompt))
• self.retrievalpipeline — composed retrieval pipeline (use await self.retrievalpipeline.retrievebyid(queryid, top_k))
• self.service — GenerationPipelineService (use self.service.getchunkcontents(chunkids), self.getquerytext(query_id))

Must return a GenerationResult(text=...) (from autoragresearch.orm.service.generationpipeline).

DO NOT add your own asyncio.gather, asyncio.Semaphore, or any concurrency control.
The base pipeline's run() already handles parallel execution of all queries via
runwithconcurrencylimit() (semaphore + gather), controlled by the maxconcurrency
config parameter. Your _generate method is called once per single query — just implement
the retrieve-and-generate logic for that one query.

Custom parameters: Add fields to your config class and pass them via getpipelinekwargs() → accept them in the pipeline constructor.

Inherited config fields (from BaseGenerationPipelineConfig):

• llm — LLM model string (auto-converted to LangChain model instance)
• retrievalpipelinename — name of the retrieval pipeline to compose with (Executor injects it)

3. Write tests and install

Use langchaincore.languagemodels.FakeListLLM to mock the LLM in tests.

cd my_rag_plugin
pip install -e .   # or: uv pip install -e .
cd .. && autorag-research plugin sync

Verify: ls configs/pipelines/generation/my_rag.yaml

Key Files

Examples

Study these existing implementations for patterns:

• autoragresearch/pipelines/generation/basicrag.py — Simple retrieve-then-generate (start here)
• autorag_research/pipelines/generation/ircot.py — Interleaving retrieval with chain-of-thought
• autorag_research/pipelines/generation/et2rag.py — Entity-aware RAG
• autoragresearch/pipelines/generation/mainrag.py — Main RAG pipeline
• YAML configs: configs/pipelines/generation/basic_rag.yaml, configs/pipelines/generation/ircot.yaml