The Agent Skills Directory

Data Handling and External Upload: The skill zips the entire workspace (excluding common build artifacts) and uploads it to an AWS S3 bucket (s3://security-agent-scans-...). This process requires broad read access to workspace files and involves sending data to an external service. While this is the primary purpose of the skill and targets the vendor's own infrastructure, the movement of workspace data to the cloud is a relevant security consideration.
Indirect Prompt Injection Surface: The skill ingests data from the AWS Security Agent service, specifically vulnerability descriptions and remediation guidance. This external content is used to inform the agent's summaries and code modification tasks. This represents a potential surface where external data could influence the agent's behavior, although this is a common pattern for security analysis tools.
Ingestion points: Findings are fetched via aws securityagent list-findings and batch-get-findings in SKILL.md.
Boundary markers: No specific boundary markers or instructions to ignore embedded instructions are defined for the processed findings.
Capability inventory: The skill can execute shell commands, write to the filesystem (.security-agent/findings-*.md), and modify code via the agent's editing tools.
Sanitization: No explicit sanitization or validation of the external finding content is mentioned before it is processed by the agent.
Command Execution: The skill makes extensive use of the aws CLI to interact with the Security Agent service, manage jobs, and handle file uploads. These commands are used to perform the core functions of the skill and integrate with the associated cloud platform.

scanning-with-aws-security-agent