The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on the AWS CLI to manage Lambda functions, update configurations, and deploy code packages, requiring significant shell command capabilities.
[EXTERNAL_DOWNLOADS]: The skill utilizes the official AWS Lambda Python runtime image from the public ECR registry for local environment testing.
[INDIRECT_PROMPT_INJECTION]: The skill facilitates the processing of external, untrusted data which can serve as a vector for indirect prompt injection.
Ingestion points: Handlers in debugging.md for SQS and API Gateway events ingest record bodies directly from external triggers.
Boundary markers: No specific delimiters or safety instructions are provided to the agent to treat event data as untrusted or to ignore embedded instructions.
Capability inventory: The skill environment possesses high-tier capabilities including shell execution (AWS CLI), local file manipulation (zip), and AWS API interactions via boto3.
Sanitization: Examples in debugging.md demonstrate raw logging of event payloads using print() and standard logging without prior sanitization or escaping of the content.

lambda