Agent Guardrails Python Packages

solvi

A runtime where decision models propose and your code decides: typed questions, hard rules, quotes checked against the text, escalation with a guaranteed error budget, replayable traces. Works with any System One model.

sponsio

Deterministic safety solutions for probabilistic AI agents

vizier-guard

Deterministic authorization for AI agent actions: owner-signed delegation grants, ALLOW/BLOCK/REVIEW decisions, and request-bound receipts. REST, MCP, and A2A.

toolboundary

Local runtime security and policy enforcement for AI agents and LLM tool calls, with permissions, guardrails, rate limits, approvals, auditing, and an emergency kill switch.

jamjet-guardrails

Deterministic content guardrails for LLM and agent applications. Prompt injection in the encoding, PII, secrets and your own rules, with published precision and recall and provenance on every decision.

sadasend

The official Python SDK for SadaSend — high-throughput transactional email with unbypassable server-enforced guardrails designed for human software engineering teams and autonomous AI agents.

driftguard-ai

Semantic mistake-memory and guardrail for AI agents. Stops agents from repeating the same failures using causal graphs and semantic matching. Supports MCP, LangGraph, and in-process Python.

gubernaut-sdk

Gubernaut 1.0: a deterministic cognitive governor for LLM agents. A drop-in, OpenAI-compatible local proxy that hard-stops runaway agent loops. Token-free control layer, Apache-2.0, measured receipts included.

jamjet-guardrails-validators

Deterministic content guardrails for LLM and agent applications. Prompt injection in the encoding, PII, secrets and your own rules, with published precision and recall and provenance on every decision.

jamjet-guardrails-nemo

Deterministic content guardrails for LLM and agent applications. Prompt injection in the encoding, PII, secrets and your own rules, with published precision and recall and provenance on every decision.