Ryoma Nicer
A Pydantic v2–compatible fork of Ryoma AI Platform
Overview
Ryoma Nicer is a community-maintained fork of the Ryoma AI Platform, originally created by the Ryoma team. The upstream project provides an AI-powered data agent framework for seamless data analysis, engineering, and visualization.
This fork is:
- Pydantic v2–compatible (
ryoma-nicerruns on Pydantic 2.x) - LangGraph-ready, updated for modern LangGraph (
StateGraph+Pregel) APIs - Actively maintained as part of the “NICER” suite of agents (e.g. baby-NICER)
It is used in production within the NICER stack to power agents such as
WorkflowAgent and SQL agents that run on a LangGraph / LangChain / LangMem
tooling ecosystem.
For the original project, see project-ryoma/ryoma. All architectural credit for the base design goes to the upstream maintainers.
Why This Fork?
Pydantic v2 introduced breaking changes in how models are defined and validated. Many projects, including the original Ryoma AI Platform, were tightly coupled to Pydantic v1. To allow developers to adopt the latest Pydantic improvements without sacrificing Ryoma functionality, this fork:
- Migrates all core model definitions to the Pydantic v2
BaseModelAPI. - Updates validation and config patterns to leverage v2’s faster runtime and stricter type checking.
- Aligns with a modern agent stack, including:
- LangGraph (tested with
langgraph==0.6.x) - LangChain / LangChain Core (tested with
langchain==0.3.x,langchain-core==0.3.x) - Usage inside the NICER suite of agents (e.g. baby-NICER / langgraph-slack).
- LangGraph (tested with
Installation
Install the Pydantic v2–compatible release from PyPI:
pip install ryoma-nicer
Optionally, include supported data source extras:
pip install "ryoma-nicer[snowflake,pyspark,postgres,sqlite,mysql,bigquery]"
Usage
The API surface and usage mirror the original Ryoma. For example, to run a simple SQL agent:
from ryoma_ai.agent.sql import SqlAgent
from ryoma_ai.datasource.postgres import PostgresDataSource
# Initialize data source
datasource = PostgresDataSource("postgresql://user:pass@host:5432/db")
# Create and run SQL agent
agent = SqlAgent("gpt-3.5-turbo").add_datasource(datasource)
agent.stream("SELECT count(*) FROM orders", display=True)
For conceptual details and additional examples, the upstream Ryoma documentation still applies in most cases. Where this fork differs is primarily in:
- Pydantic v2 model definitions
- Updated integrations used by the NICER agent stack
Usage in the NICER agent suite
Within the NICER ecosystem (e.g. baby-NICER / langgraph-slack), this fork
powers agents that run on LangGraph’s compiled graphs (StateGraph.compile(...)
returning a Pregel runtime). For example, WorkflowAgent:
- builds a
StateGraphoverMessageState - compiles it to a
Pregelruntime withMemorySavercheckpointers - uses LangGraph’s
invoke/streamAPIs to manage tool-augmented workflows
This ensures ryoma-nicer works cleanly with modern LangGraph-based systems
while preserving the spirit of the original Ryoma design.
Contribution & Upstream
- Forked from: project-ryoma/ryoma
- Upstream credit: Architectural ideas, core abstractions, and much of the original code come from the Ryoma team.
- Issue tracker & pull requests: Please use this repository to report Pydantic v2 / modern-stack compatibility issues or to propose improvements needed by the NICER suite of agents.
License
This project is released under the Apache License 2.0.
1 maintainer
Metadata
1 maintainer
Release files for ryoma-nicer 0.9.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| ryoma_nicer-0.9.0.tar.gz | 2.6 MB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| ryoma_nicer-0.9.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 2.8 MB
Release files / ryoma_nicer-0.9.0.tar.gz
| Download URL | ryoma_nicer-0.9.0.tar.gz |
|---|---|
| Size | 2.6 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
93fb3e5861a905073706d2c171e9915685afaf723606d473b61591312146c4bc
|
|
BLAKE2b-256 checksum How to use checksums |
04305c5d5e6ed15f4991d89f7be7d928796298a77b23c298190e960aa53a144c
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 21, 2026.
Transparency logRelease files / ryoma_nicer-0.9.0-py3-none-any.whl
| Download URL | ryoma_nicer-0.9.0-py3-none-any.whl |
|---|---|
| Size | 154.1 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
22968171f3101e8a1011607c69d2648b7a8c4eec324fd5b8e0e48910b063165f
|
|
BLAKE2b-256 checksum How to use checksums |
c60771300892221b718d75b64799088866fa9209880f413b816eb5f96a56ba96
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 21, 2026.
Transparency log
