The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external system data, creating an attack surface for indirect prompt injection.
Ingestion points: Infrastructure assessment, performance analysis, and security audits involve reading external data such as system logs, cloud configurations, and vulnerability scan results as described in the Workflow Process.
Boundary markers: The instructions lack explicit delimiters or instructions to ignore embedded commands within the processed data.
Capability inventory: The skill provides templates for executing shell scripts, Terraform configurations, and AWS CLI operations across the infrastructure.
Sanitization: There are no documented procedures for sanitizing or validating external input before it is processed or incorporated into reports.
[COMMAND_EXECUTION]: The skill utilizes shell scripts and CLI tools for critical system maintenance and infrastructure provisioning.
Evidence: The SKILL.md file includes a complex bash script for automated backups and multi-resource Terraform configurations.
[DATA_EXFILTRATION]: The included backup script performs network operations to external endpoints for status notifications.
Evidence: The script uses 'curl' to send POST requests to a configurable '$NOTIFICATION_WEBHOOK'. While the content is limited to status messages in the template, it establishes an outbound network channel.

Infrastructure Maintainer