The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides patterns for processing external input through Amazon States Language (ASL) to orchestrate various AWS services. While this creates a surface for indirect injection, it is the primary function of the technology described.
Ingestion points: The input parameter in start_execution calls and the file://workflow.json configuration in CLI examples.
Boundary markers: None explicit in the examples; workflows are shown processing raw JSON paths (e.g., $.order_id).
Capability inventory: The orchestrated tasks include executing Lambda functions, writing to DynamoDB, sending SQS/SNS messages, and running ECS tasks.
Sanitization: No explicit validation or sanitization logic is included in the instructional examples, which is standard for architectural documentation.
[COMMAND_EXECUTION]: The skill utilizes official AWS CLI commands for state machine management and debugging.
Evidence: Examples include aws stepfunctions create-state-machine, aws stepfunctions get-execution-history, and aws stepfunctions validate-state-machine-definition.

step-functions