# ZenGuard — Runtime Guardrails for AI Agents

> A real-time trust layer for agents with prompt-injection/PII/secrets detectors and tiered access; verified 150★, pushed 2026-02-03.

## Install

Copy the content below into your project:

## Quick Use

```bash
pip install zenguard
export ZEN_API_KEY="your_key"
# README shows configuring Dedicated tier in Python:
python -c "import os; from zenguard import Credentials, ZenGuard, ZenGuardConfig, Tier; cfg=ZenGuardConfig(credentials=Credentials(api_key=os.environ.get('ZEN_API_KEY')), tier=Tier.DEDICATED); ZenGuard(config=cfg); print('ok')"
```

## Intro

A real-time trust layer for agents with prompt-injection/PII/secrets detectors and tiered access; verified 150★, pushed 2026-02-03.

**Best for:** Teams adding runtime safety checks (injection, PII, secrets) around tool-using agents

**Works with:** Python apps and agent frameworks; README links LangChain + LlamaIndex integrations and docs.zenguard.ai

**Setup time:** 10-25 minutes

### Key facts (verified)

- GitHub: 150 stars · 21 forks · pushed 2026-02-03.
- License: MIT · owner avatar + repo URL verified via GitHub API.
- README-backed entrypoint: `pip install zenguard`.

## Main

- Start with one detector policy per risk (prompt injection vs secrets vs PII) so you can tune false positives separately.

- Pick the tier intentionally: README describes BASE vs DEDICATED tiers; use the free tier to validate integration first.

- Wrap the agent at runtime: put checks around tool calls and retrieval inputs so the agent sees sanitized context.

- Log decisions and outcomes: guardrails are most useful when you can audit why something was blocked or redacted.

### Source-backed notes

- README lists detectors such as prompt injection, PII, banned/allowed topics, keywords, and secrets (with Colab links).
- README shows install commands `pip install zenguard` and `poetry add zenguard`.
- README states that as of 03.07.2025 multi-detector detection is supported only on the Dedicated tier.

### FAQ

- **Do I need a ZenGuard API key?**: README examples reference `ZEN_API_KEY` for credentials.
- **What tiers exist?**: README describes BASE (trial/rate-limited) and DEDICATED (higher QPS; standard/enterprise).
- **Can I use multiple detectors together?**: README says multi-detector detection is Dedicated-only as of 2025-07-03.

## Source & Thanks

> Source: https://github.com/ZenGuard-AI/fast-llm-security-guardrails
> License: MIT
> GitHub stars: 150 · forks: 21

---

<!-- ZH -->

## Quick Use

```bash
pip install zenguard
export ZEN_API_KEY="your_key"
# README shows configuring Dedicated tier in Python:
python -c "import os; from zenguard import Credentials, ZenGuard, ZenGuardConfig, Tier; cfg=ZenGuardConfig(credentials=Credentials(api_key=os.environ.get('ZEN_API_KEY')), tier=Tier.DEDICATED); ZenGuard(config=cfg); print('ok')"
```

## Intro

ZenGuard 是面向 AI agents 的运行时信任层，提供提示注入/PII/Secrets 等检测器与分层套餐，适合把“安全检查”变成可复用步骤；已验证 150★，更新于 2026-02-03。

**Best for:** 希望为工具型 agent 增加运行时安全检测（注入/隐私/密钥）的团队

**Works with:** Python 应用与常见 agent 框架；README 提供 LangChain/LlamaIndex 集成链接与 docs.zenguard.ai 文档

**Setup time:** 10-25 minutes

### Key facts (verified)

- GitHub：150 stars · 21 forks；最近更新 2026-02-03。
- 许可证：MIT；作者头像与仓库链接均已通过 GitHub API 复核。
- README 中可对照的入口命令：`pip install zenguard`。

## Main

- 按风险拆分策略：把提示注入、Secrets、PII 等检测分开配置，便于分别调阈值与误报率。

- 明确选择 tier：README 描述 BASE 与 DEDICATED；先用免费层验证接入，再决定是否升级。

- 放在运行时包住 agent：在工具调用与检索输入之前做检查，让 agent 看到“干净上下文”。

- 把决策写进日志：安全护栏的价值在于可审计，能回溯为何被拦截或被脱敏。

### Source-backed notes

- README 列出检测器方向：提示注入、PII、allowed/banned topics、keywords、secrets，并提供 Colab 示例链接。
- README 给出安装命令：`pip install zenguard` 与 `poetry add zenguard`。
- README 写明：截至 2025-07-03，多检测器联合检测仅在 Dedicated tier 支持。

### FAQ

- **需要 ZenGuard 的 API key 吗？**：README 示例使用 `ZEN_API_KEY` 作为凭证。
- **有哪些 tier？**：README 描述 BASE（试用/限速）与 DEDICATED（更高 QPS；标准/企业）。
- **能同时跑多个检测器吗？**：README 表示截至 2025-07-03，多检测器检测仅 Dedicated 支持。

## Source & Thanks

> Source: https://github.com/ZenGuard-AI/fast-llm-security-guardrails
> License: MIT
> GitHub stars: 150 · forks: 21


---
Source: https://tokrepo.com/en/workflows/zenguard-runtime-guardrails-for-ai-agents
Author: Agent Toolkit