[![Paper][paper-shield]][paper-url] [![Document][document-shield]][document-url] [![Contributors][contributors-shield]][contributors-url] [![Forks][forks-shield]][forks-url] [![Stargazers][stars-shield]][stars-url] [![Issues][issues-shield]][issues-url] [![MIT License][license-shield]][license-url] [![LinkedIn][linkedin-shield]][linkedin-url]

<br /> <div align="center"> <a href="https://kyotoai.org"> <img src="images/seimei_architecture.png" alt="Logo" width="640" height="360"> </a> <h3 align="center">SEIMEI</h3> <p align="center"> <strong>S</strong>earch-<strong>E</strong>nhanced <strong>I</strong>nterface for <strong>M</strong>ulti-<strong>E</strong>xpertise <strong>I</strong>ntegration </p> <p align="center"> Unlike conventional RL that only optimizes knowledge inside the LLM, SEIMEI jointly optimizes external knowledge, enabling AI to truly absorb domain-specific and tacit expertise. Build much more personalized AI trained only for you with dramatically lower cost and higher adaptability!! <br /> <a href="https://github.com/kyotoai/SEIMEI/tree/main/demo"><strong>Explore the docs »</strong></a> <br /> <br /> <a href="https://github.com/kyotoai/SEIMEI/tree/main/demo">View Demo</a> · <a href="https://github.com/kyotoai/SEIMEI/issues/new?labels=bug&template=bug-report---.md">Report Bug</a> · <a href="https://github.com/kyotoai/SEIMEI/issues/new?labels=enhancement&template=feature-request---.md">Request Feature</a> </p> </div>  <details> <summary>Table of Contents</summary> <ol> <li> <a href="#about-the-project">About The Project</a> </li> <li> <a href="#quick-start">Quick Start</a> <ul> <li><a href="#installation">Installation</a></li> <li><a href="#set-api-key">Set API key</a></li> <li><a href="#run-seimei">Run SEIMEI</a></li> </ul> </li> <li> <a href="#usage-a-integrate-your-own-knowledge">Usage A. Integrate your own knowledge</a> <ul> <li><a href="#1-prepare-your-knowledge-file">1. Prepare your knowledge file</a></li> <li><a href="#2-run-seimei-with-knowledge-loading">2. Run SEIMEI with knowledge loading</a></li> <li><a href="#3-automatic-knowledge-accumulation-optional">3. Automatic knowledge accumulation (optional)</a></li> </ul> </li> <li> <a href="#usage-b-train-reward-model-to-optimize-knowledge">Usage B. Train Reward Model To Optimize Knowledge</a> <ul> <li><a href="#1-inferences-sampling">1. Inferences sampling</a></li> <li><a href="#2-data-conversion">2. Data conversion</a></li> <li><a href="#3-train-reward-model">3. Train reward model</a></li> <li><a href="#4-evaluate-your-model">4. Evaluate your model</a></li> </ul> </li> <li> <a href="#usage-c-cli-chat">Usage C. CLI Chat</a> </li> <li><a href="#contributing">Contributing</a></li> <li><a href="#license">License</a></li> <li><a href="#contact">Contact</a></li> <li><a href="#acknowledgments">Acknowledgments</a></li> </ol> </details>

About The Project

Search The Best Knowledge For Accurate Thought

<br /> <div align="center"> <img src="images/seimei_idea.png" alt="seimei" width="640" height="400"> </div> <br /> Here's the example of how SEIMEI works. Each agent interacts with LLM and document and makes inference. These inferences are automatically integrated by search engine and gives an answer of question. <p align="right">(<a href="#readme-top">back to top</a>)</p>

Quick Start

Installation

You can install SEIMEI using git clone the library

git clone https://github.com/kyotoai/SEIMEI.git
pip install -e SEIMEI/

Set API key

Get KyotoAI API key from https://kyotoai.net
Run

export KYOTOAI_API_KEY="(your_kyotoai_api_key)"

Run SEIMEI

In CLI app

Open seimei terminal app inside your project directory by

seimei

and start asking question.

python code

import asyncio
from seimei import seimei

async def demo_code_act():
    orchestrator = seimei(
        llm_config={"model": "gpt-5-nano"},
        max_tokens_per_question=30000,
    )

    result = await orchestrator(
        messages=[
            {"role": "user", "content": "Analyze the current directory and change."},
        ],
    )

asyncio.run(demo_code_act())

Usage A. Integrate your own knowledge

Overview

Prepare a knowledge file.
Create a CSV with reusable hints for each agent (think, code_act, answer, web_search, or * for all agents).
This becomes your portable memory layer that can be reused across runs.
Run SEIMEI with knowledge loading rules.
Pass knowledge_load_config to load CSV/JSON/JSONL files and to inject inline, step-specific hints.
This lets you control both what knowledge is injected and when it is used.
(Optional) Accumulate new knowledge automatically.
Enable knowledge_generate_config to append run retrospectives into your CSV after each run.
The newly generated rows are returned in the response and immediately reusable.

1. Prepare your knowledge file

Create seimei_knowledge/knowledge.csv (minimum columns: agent, knowledge).

agent,knowledge,tags,step,id
code_act,"Prefer rg before grep when scanning large repos","[\"search\",\"shell\"]",,101
think,"Before choosing next action, summarize the last 2 agent findings in one sentence","[\"planning\"]",">=2",102
answer,"End with a short numbered next-step list when uncertainty remains","[\"response\"]",,103
*,"Always verify file paths before proposing edits","[\"safety\"]",,104

agent: target agent name (* means all agents).
knowledge: guidance text injected into that agent.
tags (optional): JSON list or comma-separated string.
step (optional): step constraint like 2, >=2, <4, or >=1,<=3.
id (optional): stable identifier for tracking and updates.

You can also bootstrap entries with the built-in generator:

python3 -m seimei.knowledge.generate_from_generators \
  --count 25 \
  --output seimei_knowledge/knowledge.csv

2. Run SEIMEI with knowledge loading

import asyncio
from seimei import seimei

async def main():
    orchestrator = seimei(
        llm_config={"model": "gpt-5-nano"},
        allow_code_exec=True,
        max_tokens_per_question=30000,
    )

    result = await orchestrator(
        messages=[
            {"role": "user", "content": "Inspect this repo and suggest a safe cleanup plan."},
        ],
        knowledge_load_config=[
            {"load_knowledge_path": "seimei_knowledge/knowledge.csv"},
            {
                "step": [1, 2],
                "agent": "code_act",
                "text": "Run read-only commands first (pwd, ls, rg) before any edits.",
                "tags": ["safety", "planning"],
            },
            {
                "step": 3,
                "agent": ["think", "answer"],
                "text": "Explicitly list unresolved uncertainties before finalizing.",
                "tags": ["quality"],
            },
        ],
    )
    print(result["output"])

asyncio.run(main())

3. Automatic knowledge accumulation (optional)

Provide knowledge_generate_config when calling the orchestrator to append run retrospectives into a CSV knowledge base:

result = await orchestrator(
    messages=[{"role": "user", "content": "Find clever ways to speed up our ETL pipeline."}],
    knowledge_generate_config={

SEIMEI

Install / Use

README