The Agent Skills Directory

Command Execution: The skill contains detailed instructions for managing AWS Route 53 resources via the AWS CLI and MCP server. These operations are the primary function of the skill and are accompanied by consistent recommendations to use ephemeral, least-privilege IAM credentials (e.g., via IAM roles or SSO) and read-only credentials for inspection tasks.
Indirect Prompt Injection surface: The skill ingests external data such as domain names, VPC IDs, and IP addresses, which are then interpolated into CLI commands. This represents a potential attack surface for indirect prompt injection. However, the skill incorporates several mitigating controls.
Ingestion points: User-provided parameters defined in the Procedure sections of the reference files (e.g., custom_domains, vpc_ids, on_prem_domains).
Boundary markers: The instructions explicitly require the agent to ask for all parameters upfront, confirm full configurations with the customer before any write operations, and verify the successful completion of each step.
Capability inventory: Extensive write capabilities for Route 53 including record modification, firewall rule creation, and resolver endpoint configuration across multiple scripts.
Sanitization: The skill relies on the agent's instructions to confirm input with the user and the inherent validation performed by the AWS API.
Security Best Practices: The skill consistently mandates high security standards for the operations it performs. This includes requirements to scope security group rules for port 53, enable Route 53 query logging to encrypted destinations (KMS/SSE), and utilize AWS Secrets Manager for sensitive access tokens.

route53