The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the gcloud and gsutil command-line interfaces to inspect cloud resources, audit IAM policies, and manage project lifecycles. These operations are conducted locally through the agent's shell and are necessary for the skill's stated administrative purpose.
[PROMPT_INJECTION]: The skill contains a 'Safety Guidelines' section and a multi-phase workflow that enforces human-in-the-loop confirmation. It instructs the agent not to execute deletion commands during the discovery phase and to confirm each target individually with the user, which serves as a defensive measure against accidental or unauthorized project destruction.
[SAFE]: No evidence of malicious obfuscation, credential harvesting, or unauthorized data exfiltration was found. The skill references official Google Cloud Console URLs and follows standard operational procedures for cloud management.

gcp-project-cleanup