Source: https://seerepo.com/r/OpenSPG/KAG/built
Markdown source: https://seerepo.com/r/OpenSPG/KAG/built.md
Description: KAG is a Python repo built with FastAPI. It uses AI through dashscope, nltk, mcp, Ollama, and OpenAI. It has 100,755 written lines, with 35 test files.

# What is KAG built with?

KAG is a Python repo built with FastAPI. It uses AI through dashscope, nltk, mcp, Ollama, and OpenAI, makes its content with pdfminer.six, pypdf, jinja2, markdown, bs4, and 2 more, validates its data with Pydantic, is tested with Pytest, runs in production with aliyun-log-python-sdk and psutil, and uses pyodps, Neo4j, and diskcache. It has 100,755 written lines, with 35 test files.

Read on November 19, 2026, at commit e162ce1.

Repository: https://github.com/OpenSPG/KAG

## Frontend

### Interface

- **[matplotlib](https://seerepo.com/uses/matplotlib)**: 1 file
- **pyvis**: 1 file

## Backend

### The base

- **[FastAPI](https://seerepo.com/uses/fastapi)**

### Data

- **pyodps**
- **[Neo4j](https://seerepo.com/uses/neo4j)**: 2 files
- **[diskcache](https://seerepo.com/uses/diskcache)**: 1 file

### AI

- **[dashscope](https://seerepo.com/uses/dashscope)**
- **[nltk](https://seerepo.com/uses/nltk)**
- **[mcp](https://seerepo.com/uses/mcp)**: 2 files
- **[Ollama](https://seerepo.com/uses/ollama)**: 2 files

### Services

- **[OpenAI](https://seerepo.com/uses/openai)**: 1 file · [kag/bin/commands/builder.py, line 22](https://github.com/OpenSPG/KAG/blob/e162ce1884ae9f5d7357091f18be9d2d722c5834/kag/bin/commands/builder.py#L22)

## Support

### Content

- **[pdfminer.six](https://seerepo.com/uses/pdfminersix)**
- **[pypdf](https://seerepo.com/uses/pypdf)**
- **[jinja2](https://seerepo.com/uses/jinja2)**: 4 files
- **[markdown](https://seerepo.com/uses/markdown)**: 2 files
- **[bs4](https://seerepo.com/uses/bs4)**: 1 file
- **[pypdf2](https://seerepo.com/uses/pypdf2)**: 1 file
- **[python-docx](https://seerepo.com/uses/python-docx)**: 1 file

### Validation

- **[Pydantic](https://seerepo.com/uses/pydantic)**: 1 file

### Testing

- **Pytest**: 2 files

### Deploy

- **aliyun-log-python-sdk**
- **[psutil](https://seerepo.com/uses/psutil)**

### Tools

- **click**: 6 files
- **docstring-parser**: 1 file
- **gitpython**: 1 file

### Other

- **#**
- **charset-normalizer**
- **dateutils**
- **deprecated**
- **json5**
- **protobuf**
- **pycryptodome**
- **R**
- **retrying**
- **wget**
- **six**: 122 files
- **tenacity**: 27 files
- **NumPy**: 9 files
- **pandas**: 8 files
- **requests**: 7 files
- **tqdm**: 7 files
- **pyyaml**: 6 files
- **networkx**: 5 files
- **pyhocon**: 4 files
- **ruamel.yaml**: 3 files
- **aiolimiter**: 2 files
- **cachetools**: 2 files
- **urllib3**: 2 files
- **aiofiles**: 1 file
- **certifi**: 1 file
- **httpx**: 1 file
- **jieba**: 1 file
- **json-repair**: 1 file
- **portalocker**: 1 file
- **python-dateutil**: 1 file
- **schedule**: 1 file
- **tabulate**: 1 file

## Other repos with a similar stack

- [bojieli/ai-agent-book](https://seerepo.com/r/bojieli/ai-agent-book/built): FastAPI, OpenAI, matplotlib, nltk, MCP, Ollama, pypdf, markdown, pypdf2, python-docx, Pydantic, psutil · [Compare](https://seerepo.com/compare/bojieli/ai-agent-book/vs/OpenSPG/KAG)
- [run-llama/llama_index](https://seerepo.com/r/run-llama/llama_index/built): Neo4j, OpenAI, matplotlib, dashscope, nltk, MCP, Ollama, pypdf, jinja2, bs4, Pydantic, psutil · [Compare](https://seerepo.com/compare/OpenSPG/KAG/vs/run-llama/llama_index)
- [datawhalechina/hello-agents](https://seerepo.com/r/datawhalechina/hello-agents/built): FastAPI, OpenAI, matplotlib, MCP, pdfminer.six, pypdf, jinja2, markdown, pypdf2, python-docx, Pydantic · [Compare](https://seerepo.com/compare/datawhalechina/hello-agents/vs/OpenSPG/KAG)
- [hsliuping/TradingAgents-CN](https://seerepo.com/r/hsliuping/TradingAgents-CN/built): FastAPI, OpenAI, matplotlib, dashscope, MCP, jinja2, markdown, python-docx, Pydantic, psutil · [Compare](https://seerepo.com/compare/hsliuping/TradingAgents-CN/vs/OpenSPG/KAG)
- [HKUDS/Vibe-Trading](https://seerepo.com/r/HKUDS/Vibe-Trading/built): FastAPI, OpenAI, matplotlib, MCP, pypdf, jinja2, bs4, python-docx, Pydantic · [Compare](https://seerepo.com/compare/HKUDS/Vibe-Trading/vs/OpenSPG/KAG)
- [Significant-Gravitas/AutoGPT](https://seerepo.com/r/Significant-Gravitas/AutoGPT/built): FastAPI, OpenAI, Ollama, pypdf, jinja2, python-docx, Pydantic, psutil · [Compare](https://seerepo.com/compare/OpenSPG/KAG/vs/Significant-Gravitas/AutoGPT)

## Frequently asked questions

### What language is KAG written in?

KAG is written in Python, with 100,755 written lines.

### What framework does KAG use?

KAG is built with FastAPI.

### Does KAG use AI?

Yes. KAG uses AI through dashscope, nltk, mcp, Ollama, and OpenAI.

### What services does KAG use?

KAG uses pyodps, Neo4j, and diskcache.

### Does KAG have tests?

Yes. KAG has 35 test files and is tested with Pytest.

### Which repos have a stack like KAG's?

The ones that share most of its stack: bojieli/ai-agent-book, run-llama/llama_index, and datawhalechina/hello-agents.

## What is your repo built with?

Any public repo, with no account: https://seerepo.com/r/owner/repo/built, with the owner and name it has on GitHub.
