Skip to content
System One

CryptoOda/typesafe-guard

LLM-agnostic guardrails library using Jev to detect prompt injection, jailbreak attempts, PII exposure and credential leaks in one batched request. Not published to PyPI at time of check.

CryptoOda/typesafe-guard on GitHub

Stars
1
Forks
0
Language
Python
License
MIT
Last push
17 Sep 2026

Read from GitHub on . Counts change daily.

How it uses Jev

def scan(self, text: str, *, checks: Mapping[str, Noul] | None = None) -> ScanResult:
    """Run all configured checks over `text` in a single batched request."""
    active = dict(checks) if checks is not None else self._checks
    response = self._client.system_one(state=text, questions=active)
    results = {
        name: CheckResult(name=name, probability=answer.noul)
        for name, answer in response.nouls.items()
    }
    return ScanResult(
        results=results,
        input_tokens=response.usage.input_tokens,
        output_tokens=response.usage.output_tokens,
    )

View in typesafe_guard/core.py

Open on GitHub

More like this

Use cases this is tagged with