The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified in the skill instructions or referenced documentation. The content correctly identifies and discourages common security anti-patterns like public S3 buckets, long-lived access keys, and hardcoded credentials.
[COMMAND_EXECUTION]: The skill includes a local utility script scripts/verify.sh. This script is a read-only linter that scans local configuration files for security vulnerabilities using standard command-line tools like grep and find. It does not perform network requests or modify any files.
[PROMPT_INJECTION]: The skill contains a surface for indirect prompt injection as it facilitates the generation of AWS IAM policies and configurations based on user input. This is a low-risk architectural characteristic rather than a malicious intent; the risk is mitigated by the skill's guidance to use AWS Access Analyzer and the provided verify.sh tool to validate all generated policies before deployment.

aws-essentials