aws-serverless

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect prompt injection surface detected.\n
  • Ingestion points: The skill processes untrusted data from event.body and SQS event.Records in SKILL.md.\n
  • Boundary markers: No delimiters or instructions to ignore embedded commands are used when handling external event data.\n
  • Capability inventory: The code snippets perform database writes and reads (docClient.send, saveToDatabase) based on the content of the ingested event data.\n
  • Sanitization: The logic lacks input validation or sanitization before processing the parsed JSON data.\n- [SAFE]: The skill uses official and trusted AWS SDKs for Node.js (@aws-sdk/*) and Python (boto3).\n- [SAFE]: SAM templates demonstrate best practices by using resource-scoped IAM policies (DynamoDBReadPolicy, DynamoDBCrudPolicy) rather than wildcard permissions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 09:57 PM
Security Audit — agent-trust-hub — aws-serverless