The Agent Skills Directory

[DATA_EXPOSURE_&_EXFILTRATION]: The skill utilizes environment variables (e.g., process.env.TABLE_NAME) for resource configuration, which is a standard AWS best practice. While code snippets include error logging that outputs stack traces to console.error, this is typical for development templates and does not constitute a security violation in this context.
[INDIRECT_PROMPT_INJECTION]: The provided Lambda handlers ingest untrusted data from event.body and SQS record.body.
Ingestion points: handler.js and processor.js parse JSON from event bodies.
Boundary markers: None present in the code snippets.
Capability inventory: The templates demonstrate DynamoDB read/write operations and SQS message processing.
Sanitization: The snippets show basic JSON.parse usage without explicit schema validation, which is an expected starting point for template code. This represents a typical attack surface for serverless functions rather than a malicious intent within the skill itself.
[EXTERNAL_DOWNLOADS]: The skill references standard AWS SDKs for Node.js and Python. These are well-known, trusted libraries provided by Amazon Web Services and are necessary for the skill's stated purpose.

aws-serverless