ar-guardrail-deployer
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill facilitates the deployment of Amazon Bedrock Automated Reasoning policies into production environments. All operations utilize the official AWS SDK (boto3) to interact with Bedrock APIs for versioning and configuration.
- [COMMAND_EXECUTION]: The instructions in SKILL.md direct the agent to execute specific Python scripts provided within the skill package using the uv tool. These scripts, create_version.py and deploy_guardrail.py, are dedicated to legitimate AWS infrastructure management.
- [INDIRECT_PROMPT_INJECTION]: The skill accepts parameters such as policy ARNs and guardrail names from the user context. These inputs are used strictly as identifiers for API calls and are not processed as executable code or natural language instructions by the scripts, minimizing the risk of injection attacks.
Audit Metadata