---
title: "Safeguard your agentic AI applications with the Amazon Bedrock Guardrails InvokeGuardrailChecks API"
slug: "safeguard-your-agentic-ai-applications-with-the-amazon-bedrock-guardrails-invokeguardrailchecks-api"
date: 2026-06-16
category: tech-pub
tags: [agents, ai-safety, amazon]
language: en
sources_count: 1
featured: false
publisher: AInauten News
url: https://news.ainauten.com/en/story/safeguard-your-agentic-ai-applications-with-the-amazon-bedrock-guardrails-invokeguardrailchecks-api
---

# Safeguard your agentic AI applications with the Amazon Bedrock Guardrails InvokeGuardrailChecks API

**Published**: 2026-06-16 | **Category**: tech-pub | **Sources**: 1

---

## TL;DR

- AWS introduced the InvokeGuardrailChecks API for Amazon Bedrock Guardrails on June 16, 2026. It lets teams run individual safety checks inside agent workflows without creating guardrail resources or managing versions.

---

## Summary

- AWS introduced the InvokeGuardrailChecks API for Amazon Bedrock Guardrails on June 16, 2026. It lets teams run individual safety checks inside agent workflows without creating guardrail resources or managing versions.
- The API is detect-only: it does not block, mask, or rewrite content. It returns scores, then the application decides whether to block, retry, escalate, allow, or log the result.
- Supported checks include content filters, prompt attack detection for jailbreaks, prompt injection, and prompt leakage, plus sensitive-information checks across 31 PII types with text offsets.
- AWS frames it as the agentic counterpart to ApplyGuardrail: targeted checks per step, while ApplyGuardrail remains the simpler fit for uniform enforcement in request-response apps.

---

## Why it matters

AWS introduced the InvokeGuardrailChecks API for Amazon Bedrock Guardrails on June 16, 2026. It lets teams run individual safety checks inside agent workflows without creating guardrail resources or managing versions.

---

## Key Points

- AWS introduced the InvokeGuardrailChecks API for Amazon Bedrock Guardrails on June 16, 2026. It lets teams run individual safety checks inside agent workflows without creating guardrail resources or managing versions.
- The API is detect-only: it does not block, mask, or rewrite content. It returns scores, then the application decides whether to block, retry, escalate, allow, or log the result.
- Supported checks include content filters, prompt attack detection for jailbreaks, prompt injection, and prompt leakage, plus sensitive-information checks across 31 PII types with text offsets.
- AWS frames it as the agentic counterpart to ApplyGuardrail: targeted checks per step, while ApplyGuardrail remains the simpler fit for uniform enforcement in request-response apps.

---

## Nauti's Take

This is more useful than adding one giant shield over the whole app. Agents usually get messy at the handoffs: a tool returns unsafe output, a user tries prompt injection, or an intermediate step leaks PII. AWS is selling infrastructure here, and the PR gloss is visible, but the real safety work stays with builders. Scores without clear actions become better logs, not safer agents.

---


## FAQ

**Q:** What is Safeguard your agentic AI applications with the Amazon Bedrock Guardrails InvokeGuardrailChecks API about?

**A:** - AWS introduced the InvokeGuardrailChecks API for Amazon Bedrock Guardrails on June 16, 2026. It lets teams run individual safety checks inside agent workflows without creating guardrail resources or managing versions.

**Q:** Why does it matter?

**A:** AWS introduced the InvokeGuardrailChecks API for Amazon Bedrock Guardrails on June 16, 2026. It lets teams run individual safety checks inside agent workflows without creating guardrail resources or managing versions.

**Q:** What are the key takeaways?

**A:** AWS introduced the InvokeGuardrailChecks API for Amazon Bedrock Guardrails on June 16, 2026. It lets teams run individual safety checks inside agent workflows without creating guardrail resources or managing versions.. The API is detect-only: it does not block, mask, or rewrite content. It returns scores, then the application decides whether to block, retry, escalate, allow, or log the result.. Supported checks include content filters, prompt attack detection for jailbreaks, prompt injection, and prompt leakage, plus sensitive-information checks across 31 PII types with text offsets.

---

## Related Topics

- [agents](https://news.ainauten.com/en/tag/agents)
- [ai-safety](https://news.ainauten.com/en/tag/ai-safety)
- [amazon](https://news.ainauten.com/en/tag/amazon)

---

## Sources

- [Safeguard your agentic AI applications with the Amazon Bedrock Guardrails InvokeGuardrailChecks API](https://aws.amazon.com/blogs/machine-learning/safeguard-your-agentic-ai-applications-with-the-amazon-bedrock-guardrails-invokeguardrailchecks-api/) - AWS Machine Learning Blog

---

## About This Article

This article is a synthesis of 1 sources, curated and summarized by AInauten News. We aggregate AI news from trusted sources and provide bilingual (German/English) coverage.

**Publisher**: [AInauten](https://www.ainauten.com) | **Site**: [news.ainauten.com](https://news.ainauten.com)

---

*Last Updated: 2026-06-17*