lambda-handler-pattern
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized data exfiltration were detected. The skill provides reference architecture for serverless development using official AWS libraries.
- [INDIRECT_PROMPT_INJECTION]: The handler pattern involves ingesting untrusted data from an external API event, which represents a standard vulnerability surface for indirect injection.
- Ingestion points: The code parses the
event.bodyfrom anAPIGatewayProxyEventinsideSKILL.mdexamples. - Boundary markers: Data is mapped to structured JSON object keys (
id,data,timestamp) but the content within thedatafield is not further delimited or restricted. - Capability inventory: The skill demonstrates writing to a DynamoDB table using the
PutCommandfrom the@aws-sdk/lib-dynamodblibrary. - Sanitization: Input is parsed via
JSON.parse(), but no specific content filtering or instruction-ignoring delimiters are implemented for the business logic payload.
Audit Metadata