syntheticcfo - Python client

Generate forensic-grade synthetic ERP data (SAP ECC, SAP S/4HANA, Oracle Cloud) from a script or notebook. Every dataset ships with a complete labelled fraud answer key, articulated financial statements and a reproducibility certificate: the same seed and configuration reproduce the package byte for byte. No language model ever generates the numbers.

The client has no dependencies beyond the standard library. Install it straight from the platform (this is version 0.2.0, which carries the one-line command):

pip install https://app.syntheticcfo.com/sdk/syntheticcfo-0.2.0-py3-none-any.whl

The package index copy (pip install syntheticcfo) is the earlier 0.1.0 until 0.2.0 is published there; it has the same client but no command line. Or, since it is one file, take it directly and drop it anywhere:

curl -O https://app.syntheticcfo.com/sdk/syntheticcfo.py

Quickstart

Create an API key at https://app.syntheticcfo.com under Account, API access. The key carries your plan and monthly allowance. It cannot manage your account, read your chats or create more keys.

from syntheticcfo import Client

c = Client(api_key="scfo_...")            # or set the SCFO_API_KEY env var
print(c.me()["plan"])                     # your plan and remaining generations

job = c.generate(
    platform="ORACLE",                    # SAP_ECC | SAP_S4HANA | ORACLE
    industry="RETAIL",                    # TMT PHARMA MANUFACTURING FINANCIAL_SERVICES RETAIL
    fraud="high",                         # none | low | medium | high
    rows=600,
    modules=["P2P", "O2C"],               # subset of P2P O2C CE R2R HR MM FA
    jurisdiction="ZA",
    seed=42,
    wait=True,                            # poll until the package is built
)

package = job.download("out/")            # the four-layer audit package (zip)
lab = job.lab_export("out/")              # JSONL tables + answer_key.jsonl + dataset card
print(job.insights())                     # what was planted, by control

For model evaluation, add evaluation_mode=True: the shipped database then withholds the ground-truth marker columns so nothing in the data betrays the answer key, while the labels still ship separately in 01_Audit_Evidence.

Notes

• Plan limits apply exactly as in the app: over-plan requests are refused with an honest message, and row counts above your plan are clamped with a note in the package, never silently.
• A failed or interrupted generation refunds the metered credit automatically.
• Free-plan datasets and the public sample may not be redistributed, resold, or used to train or evaluate a competing product (see the Terms in the app).