bedrock-agentcore-deployment
Warn
Audited by Snyk on Feb 15, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.80). The skill's runtime examples clearly ingest and interpret arbitrary user-provided prompts (e.g., payload.get("prompt") in main.py and the agentcore invoke / curl examples), exposing the agent to untrusted user-generated content that could enable indirect prompt injection.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.80). The skill's runtime artifact references (e.g., s3://my-bucket/agents/v1.0.0/package.zip and the container image 123456789012.dkr.ecr.us-east-1.amazonaws.com/my-agent:v1.0.0) are supplied to create_agent_runtime and will be fetched and executed as the agent runtime, meaning remote content controls executed code.
Audit Metadata