AWS Security ChangesHomeSearch

AWS wellarchitected: Add AGENTSEC04 goal-alignment best-practice links

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

File: wellarchitected/latest/agentic-ai-lens/agentsec04.md

Summary

Adds two links to AGENTSEC04 best-practice pages on guardrails/alignment controls and human-in-the-loop for critical decisions.

Security assessment

Guardrails and human-in-the-loop approval are security/alignment controls for preventing agent manipulation, so this is security documentation; no specific vulnerability or CVE is referenced.

Evidence

+[AGENTSEC04-BP01 Implement guardrails and alignment controls](./agentsec04-bp01.html)  

Diff

diff --git a/wellarchitected/latest/agentic-ai-lens/agentsec04.md b/wellarchitected/latest/agentic-ai-lens/agentsec04.md
index 70ba063..4f333ea 100644
--- a//wellarchitected/latest/agentic-ai-lens/agentsec04.md
+++ b//wellarchitected/latest/agentic-ai-lens/agentsec04.md
@@ -14,0 +15,2 @@ AGENTSEC04: How do you support agent goal alignment and prevent manipulation?
+[AGENTSEC04-BP01 Implement guardrails and alignment controls](./agentsec04-bp01.html)  
+[AGENTSEC04-BP02 Human-in-the-loop for critical decisions](./agentsec04-bp02.html)