pp-workspace-admin

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from Google Workspace, including email content, Drive file metadata, and security alerts, which could contain malicious instructions designed to influence the agent's behavior.
  • Ingestion points: Untrusted data enters the context through commands like 'messages list', 'files list', 'alerts list', and 'audit external-shares'.
  • Capability inventory: The skill allows sensitive administrative actions such as 'directory-users-delete', 'directory-tokens-delete', 'mobiledevices-action' (remote wipe), and 'messages send'.
  • Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded content within the ingested Workspace data.
  • Sanitization: No explicit sanitization or filtering of external content is mentioned before the data is processed by the agent.
  • [DATA_EXFILTRATION]: The CLI tool supports a '--deliver webhook:' flag, which allows the agent to POST command results to an external URL. While useful for automation, this provides a mechanism to transmit sensitive audit data to external endpoints.
  • [EXTERNAL_DOWNLOADS]: The skill requires downloading and installing the 'workspace-admin-pp-cli' and 'workspace-admin-pp-mcp' tools via 'npx' from NPM and 'go install' from GitHub. These resources are hosted in the vendor's official repositories (mvanhorn).
  • [COMMAND_EXECUTION]: The skill operates by executing shell commands using the 'workspace-admin-pp-cli' binary to manage and audit Google Workspace environments.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 03:29 PM
Security Audit — agent-trust-hub — pp-workspace-admin