Gws Shared
Security checks across static analysis, malware telemetry, and agentic risk
Overview
This instruction-only skill coherently documents use of the gws CLI, including expected Google authentication and broad API flags, with no evidence of hidden or deceptive behavior.
Before using this skill, make sure the gws binary is one you trust, authenticate with the least-privileged Google identity that can complete the task, and require confirmation for any command that writes, deletes, uploads, or saves files.
Static analysis
No static analysis findings were reported for this release.
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
If installed and used, gws commands may act with the permissions of the logged-in Google account or service account.
The skill documents OAuth and service-account authentication, which is expected for Google Workspace CLI use but delegates account or workspace authority to subsequent commands.
# Browser-based OAuth (interactive) gws auth login # Service Account export GOOGLE_APPLICATION_CREDENTIALS=/path/to/key.json
Use the least-privileged Google account or service account needed, protect service-account key files, and review the exact target and scope before approving any write/delete action.
Incorrect or overbroad commands could modify workspace data or upload/download files if a user approves them.
The reference exposes broad CLI patterns for making Google Workspace API requests, including request bodies, file uploads, and file outputs; this is purpose-aligned, and the same artifact tells agents to confirm write/delete commands.
gws <service> <resource> [sub-resource] <method> [flags]
...
`--json '{"key": "val"}'` | Request body |
`-o, --output <PATH>` | Save binary responses to file |
`--upload <PATH>` | Upload file content (multipart) |Ask the agent to show the exact gws command, target resource, request body, and file paths before running mutating commands; use --dry-run where supported.
