aws-sdk-java-v2-bedrock

Pass

Audited by Gen Agent Trust Hub on Feb 23, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Detected an indirect prompt injection surface area because the skill facilitates processing untrusted user data through foundation models.
  • Ingestion points: User-provided prompt strings are passed directly to model invocation methods in SKILL.md and advanced-model-patterns.md.
  • Boundary markers: The documentation suggests sanitizing inputs as a best practice in the SKILL.md warnings section, but the provided code snippets do not implement automated delimiters or ignore instructions.
  • Capability inventory: The skill uses BedrockRuntimeClient for network-based model invocation and lists Bash as an allowed tool in its metadata.
  • Sanitization: No programmatic sanitization or validation logic is included in the Java patterns to filter or escape the input data before it reaches the model provider.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 23, 2026, 11:25 PM