Quick Start

Hello Experience Maker

Here is a simple user guide for ExperienceMaker.

Step0: Preparation Work

Prepare LLM & EMBEDDING_MODEL

We need to prepare the API services for the LLM and the Embedding model. Since we are using an OpenAI-compatible service, we only need to write the OPENAI_API_KEY and OPENAI_BASE_URL into the environment.

export OPENAI_API_KEY="sk-xxx"
export OPENAI_BASE_URL="xxx"

Prepare Vector Store

If you want to use vector store, you need to set up a vector database. Don't forget to set up the ES_HOSTS.

• Elasticsearch quick start

Prepare Your Own Agent

Assume you have a runnable agent. Here, we use a basic LLM combined with a simple react framework including three tools(code, web_search, terminate) as an example.

class YourOwnAgent(...):
    ...
    def think(self, **kwargs) -> bool:
        ...

    def act(self, **kwargs):
        ...    

    def run(self, query: str):
        ...

Here is an example code with prompt. To use this simple agent, you will need to set up DASHSCOPE_API_KEY.

export DASHSCOPE_API_KEY="sk-xxx"

Step1: Start ExperienceMaker Http Service

• Install dependencies.

pip install .

• We start our context and summary services in simple mode. Next, you can use standard HTTP interfaces to call the services.

python -m experiencemaker.em_service \
 --port=8001 \
 --llm='{"backend": "openai_compatible", "model_name": "qwen3-32b", "temperature": 0.6}' \
 --embedding_model='{"backend": "openai_compatible", "model_name": "text-embedding-v4", "dimensions": 1024}' \
 --vector_store='{"backend": "elasticsearch"}' \
 --context_generator='{"backend": "simple"}' \
 --summarizer='{"backend": "simple"}'

• You can also use the more comprehensive StepSummarizer and StepContextGenerator, which provide more fine-grained functionality for processing trajectories (for more comprehensive documentation, see: cookbook/step_agent/readme.md)

python -m experiencemaker.em_service \
--port=8001 \
--llm='{"backend": "openai_compatible", "model_name": "qwen-max-2025-01-25", "temperature": 0.6}' \
--embedding_model='{"backend": "openai_compatible", "model_name": "text-embedding-v4", "dimensions": 1024}' \
--vector_store='{"backend": "elasticsearch"}' \
--agent_wrapper='{"backend": "simple"}' \
--context_generator='{"backend": "step", "enable_llm_rerank": true, "enable_context_rewrite": true, "enable_score_filter": false, "vector_retrieve_top_k": 15, "final_top_k": 5, "min_score_threshold": 0.3}' \
--summarizer='{"backend": "step", "enable_step_segmentation": false, "enable_similar_comparison": false, "enable_experience_validation": true, "max_retries": 3,"max_workers":16}'

# Parameter Description

## StepSummarizer
### enable_step_segmentation: Segment trajectories into meaningful step sequences
### enable_similar_comparison: Compare similar sequences between success/failure
### enable_experience_validation: Validate experience quality before storage

## StepContextGenerator
### enable_llm_rerank: Rerank retrieved experiences by relevance using LLM
### enable_context_rewrite: Rewrite context to be more task-specific
### enable_score_filter: Filter experiences by quality scores

Step2: Enhance Your Own Agent

Call the capabilities of ContextGenerator and Summarizer through EMClient.

from experiencemaker.em_client import EMClient

em_client = EMClient(base_url="http://0.0.0.0:8001")

Assume you have a list of messages generated by an agent; you can use the summarizer to generate experiences from them.

request = SummarizerRequest(trajectories=[Trajectory(query=query, steps=messages, answer=messages[-1].content, done=True)], workspace_id="w_1234")
response = em_client.call_summarizer(request)
for experience in response.experiences:
    print(experience.model_dump_json())

Assume you have a query; you can use the context generator to generate a new query from the query and the related experiences.

request = ContextGeneratorRequest(trajectory=Trajectory(query=query), retrieve_top_k=1, workspace_id="w_1234")
response = em_client.call_context_generator(request)
new_query = f"{response.context_msg.content}\n\nUser Question\n{query}"

The complete code can be found in the implementation of YourOwnAgentEnhanced.