The Agent Skills Directory

[SAFE]: No malicious patterns, data exfiltration risks, or prompt injection attempts were detected within the skill instructions or reference files.
[REMOTE_CODE_EXECUTION]: The skill demonstrates positive security awareness by explicitly instructing developers NEVER to include Bash(curl:*) in the allowed-tools field to prevent remote code execution vulnerabilities.
[COMMAND_EXECUTION]: The provided CLI examples (e.g., Rover CLI) are standard development commands used for graph management and do not involve privilege escalation or suspicious behavior.
[DATA_EXFILTRATION]: All network references point to official Apollo GraphQL documentation or the Agent Skills specification site. No hardcoded credentials or sensitive file path access patterns were found.

skill-creator