AWS prescriptive-guidance medium security documentation change

Service: prescriptive-guidance · 2026-07-10 · Security-related medium

File: prescriptive-guidance/latest/llm-prompt-engineering-best-practices/best-practices.md

Summary

Replaced angle brackets with HTML entities in tags and updated section header from 'Common attacks' to 'Common prompt injection attacks'.

Security assessment

The change specifically highlights prompt injection attacks as a security threat and references protective tagging strategies. This documents security vulnerabilities in LLM systems and countermeasures.

Diff

diff --git a/prescriptive-guidance/latest/llm-prompt-engineering-best-practices/best-practices.md b/prescriptive-guidance/latest/llm-prompt-engineering-best-practices/best-practices.md
index 15e8f3c9b..e20042903 100644
--- a//prescriptive-guidance/latest/llm-prompt-engineering-best-practices/best-practices.md
+++ b//prescriptive-guidance/latest/llm-prompt-engineering-best-practices/best-practices.md
@@ -7 +7 @@
-Use <thinking> and <answer> tagsUse guardrails
+Use &lt;thinking&gt; and &lt;answer&gt; tagsUse guardrails
@@ -13 +13 @@ The following guardrails and best practices were tested on a RAG application tha
-## Use <thinking> and <answer> tags
+## Use &lt;thinking&gt; and &lt;answer&gt; tags
@@ -37 +37 @@ To use the Amazon Web Services Documentation, Javascript must be enabled. Please
-Common attacks
+Common prompt injection attacks