backup-disaster-recovery

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt includes literal secret values (example AWS access keys and passwords) and manifests that create or embed secrets in plaintext, which encourages the agent to output those secret values verbatim (high exfiltration risk).

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The skill fetches backups from the S3 bucket URI s3://my-backups at runtime (e.g., aws s3 cp "$BACKUP_BUCKET/.../kubernetes-*.yaml.gz" | gunzip | kubectl apply -f
• and piping SQL from S3 into psql), so remote content retrieved from s3://my-backups can directly execute configuration/commands on clusters/databases.