AWS Security ChangesHomeSearch

AWS wellarchitected: Add AGENTSEC08 input/output validation best-practice links

Service: wellarchitected · 2026-10-20 · Documentation medium

File: wellarchitected/latest/agentic-ai-lens/agentsec08.md

Summary

Adds two links to AGENTSEC08 best-practice pages on multi-layer input validation and prompt injection defense, and output filtering for sensitive information.

Security assessment

Prompt injection defense and output filtering for sensitive data are core agent security controls, so this adds security documentation; the diff is a link list with no referenced CVE or fixed flaw.

Evidence

+[AGENTSEC08-BP01 Multi-layer input validation and prompt injection defense](./agentsec08-bp01.html)  

Diff

diff --git a/wellarchitected/latest/agentic-ai-lens/agentsec08.md b/wellarchitected/latest/agentic-ai-lens/agentsec08.md
index f9ad152..8816e50 100644
--- a//wellarchitected/latest/agentic-ai-lens/agentsec08.md
+++ b//wellarchitected/latest/agentic-ai-lens/agentsec08.md
@@ -14,0 +15,2 @@ AGENTSEC08: How do you validate and secure agent inputs and outputs?
+[AGENTSEC08-BP01 Multi-layer input validation and prompt injection defense](./agentsec08-bp01.html)  
+[AGENTSEC08-BP02 Output filtering for sensitive information](./agentsec08-bp02.html)