feat(aws-bedrock-mantle/openai.gpt-oss-safeguard-120b): add new models [bot]

[models-bot]

Auto-generated by model-addition-agent for aws-bedrock-mantle/openai.gpt-oss-safeguard-120b.

---

Note

Low Risk
Metadata-only addition with no runtime or auth changes; pricing details may still need follow-up if region stubs are incomplete.

Overview
Registers openai.gpt-oss-safeguard-120b on the aws-bedrock-mantle provider via a new model YAML entry (auto-generated by the model-addition agent).

The config sets model: openai.gpt-oss-safeguard-120b, mode: unknown, and a costs list covering 13 AWS regions (region-only stubs, no per-token pricing fields yet).

^{Reviewed by Cursor Bugbot for commit c72e449. Bugbot is set up for automated code reviews on this repo. Configure here.}

[cursor]

Cursor Bugbot has reviewed your changes and found 2 potential issues.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit c72e449. Configure here.}

[cursor]

Cost entries lack token pricing

High Severity

Every costs list item only sets region and omits fields such as input_cost_per_token and output_cost_per_token. build-unified-json still emits this model because costs is non-empty, so consumers can treat the model as priced while lookups return no rates (often zero or missing billing).

 

^{Reviewed by Cursor Bugbot for commit c72e449. Configure here.}

[cursor]

Chat model marked unknown mode

Medium Severity

mode is set to unknown for openai.gpt-oss-safeguard-120b, while the existing aws-bedrock config for the same model id uses mode: chat with supportedModes including chat. Gateways that filter or route by mode may omit or mis-handle this mantle entry.

 

^{Reviewed by Cursor Bugbot for commit c72e449. Configure here.}