AWS bedrock: Swap feature support between bedrock-runtime and bedrock-mantle for Claude Opus 4.7
Summary
Corrects the model card's feature support matrix: Guardrails, Prompt optimization, Knowledge base, Model evaluation, Prompt management, Flows, and Agents are now listed as supported on the bedrock-runtime endpoint (previously marked unsupported), while Count tokens and Implicit Prompt Caching move to the bedrock-mantle endpoint. The two section headings were also swapped to match.
Security assessment
This is a documentation accuracy fix to a feature-availability table, not a fix for a vulnerability or incident. Guardrails is a safety control, but the change only records that it is now available on the bedrock-runtime endpoint rather than explaining or hardening any security mechanism, so it is not security documentation in the substantive sense.
Evidence
+ *  [Guardrails](guardrails.html)
Diff
diff --git a/bedrock/latest/userguide/model-card-anthropic-claude-opus-4-7.md b/bedrock/latest/userguide/model-card-anthropic-claude-opus-4-7.md index 0ab9f27..6a16dff 100644 --- a//bedrock/latest/userguide/model-card-anthropic-claude-opus-4-7.md +++ b//bedrock/latest/userguide/model-card-anthropic-claude-opus-4-7.md @@ -87 +87 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app -**Features supported using`bedrock-mantle` endpoint** +**Features supported using`bedrock-runtime` endpoint** @@ -93 +92,0 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app - *  [Implicit Prompt Caching](prompt-caching.html#prompt-caching-implicit) @@ -96 +95,7 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app - *  [Count tokens](count-tokens.html) + *  [Guardrails](guardrails.html) + *  [Prompt optimization](prompt-management-optimize.html) + *  [Knowledge base](knowledge-base.html) + *  [Model evaluation](evaluation.html) + *  [Prompt management](prompt-management.html) + *  [Flows](flows.html) + *  [Agents](agents.html) @@ -102,7 +106,0 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app - *  [Guardrails](guardrails.html) - *  [Prompt optimization](prompt-management-optimize.html) - *  [Knowledge base](knowledge-base.html) - *  [Model evaluation](evaluation.html) - *  [Prompt management](prompt-management.html) - *  [Flows](flows.html) - *  [Agents](agents.html) @@ -109,0 +108,2 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app + *  [Count tokens](count-tokens.html) + *  [Implicit Prompt Caching](prompt-caching.html#prompt-caching-implicit) @@ -113 +113 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app -**Features supported using`bedrock-runtime` endpoint** +**Features supported using`bedrock-mantle` endpoint** @@ -119 +118,0 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app - *  [Implicit Prompt Caching](prompt-caching.html#prompt-caching-implicit) @@ -122,7 +121 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app - *  [Guardrails](guardrails.html) - *  [Prompt optimization](prompt-management-optimize.html) - *  [Knowledge base](knowledge-base.html) - *  [Model evaluation](evaluation.html) - *  [Prompt management](prompt-management.html) - *  [Flows](flows.html) - *  [Agents](agents.html) + *  [Count tokens](count-tokens.html) @@ -133,0 +127,7 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app + *  [Guardrails](guardrails.html) + *  [Prompt optimization](prompt-management-optimize.html) + *  [Knowledge base](knowledge-base.html) + *  [Model evaluation](evaluation.html) + *  [Prompt management](prompt-management.html) + *  [Flows](flows.html) + *  [Agents](agents.html) @@ -135 +135 @@ Whenever possible, we recommend using the `bedrock-runtime` endpoint for new app - *  [Count tokens](count-tokens.html) + *  [Implicit Prompt Caching](prompt-caching.html#prompt-caching-implicit)