AI AWS Guardrails

Release date 2026/08/06

Feature

• Added last_message as a new optional value for config.text_source.

Bugfix

• Fixed an issue where AI Gateway guardrail plugins’ request extractors did not correctly honor the configured text_source option.

• Fixed an issue where AI Gateway guardrail plugins failed to extract prompt text from OpenAI-native requests using the input_text content type.

• Fixed an issue where the plugin ignored the configured text_source option when allow_masking was enabled, masking the entire raw request body instead.

• Fixed an issue where the plugins were prevented from working properly with Claude Code.

Release date 2026/04/07

Feature

• the ssl_verify option to validate TLS certificate when connecting to the bedrock service is now enabled by default.

Bugfix

• Allowed streaming of responses when guarding_mode was set to INPUT with allow_masking enabled.

Release date 2026/04/08

Bugfix

• Allowed streaming of responses when guarding_mode was set to INPUT with allow_masking enabled.

Release date 2025/12/18

Feature

• Enabled at-rest keyring encryption for sensitive fields in AI AWS Guardrails plugin.

• Added block reason info metrics to AWS Guardrails plugin analytics.

• added the flag ssl_verify to control certificate verification when connecting to the bedrock service.

Bugfix

• Fixed an issue where the error does not reflect the root cause when the guardrails_version did not match the expected pattern.

• Fixed an issue where the ai-aws-guardrails metrics could not be recorded.

Release date 2025/11/18

Bugfix

• Fixed an issue where the ai-aws-guardrails metrics could not be recorded.

Release date 2025/10/01

Feature

• Add support for AWS IAM role to the ai-aws-guardrails plugin.

Bugfix

• Fixed the plugin to work with multi-modal content.

Release date 2026/02/26

Bugfix

• Fixed an issue where the ai-aws-guardrails metrics could not be recorded.

Release date 2025/09/05

Bugfix

• Fixed the plugin to work with multi-modal content.

Release date 2025/07/03

Feature

• Added a new AI plugin that can protect requests and responses to/from AI LLM.

Bugfix

• Fixed an issue where the plugin did not support inspecting responses in non-stream mode.