Netlify's official skill for zero-config managed Postgres — querying from Functions, Drizzle setup, migrations and per-preview database branches.
Google Workspace CLI (gws)Skill
Summary
Google's official skill for the gws CLI — drive Gmail, Drive, Calendar, Sheets, Docs, Chat and Admin APIs from an agent, with Model Armor screening.
Features
- One CLI across Drive, Gmail, Calendar, Sheets, Docs, Chat and Admin
- Command surface built dynamically from Google's Discovery Service
- 100+ SKILL.md files — one per API, plus helpers like gmail +triage and +reply
- Model Armor response screening via --sanitize to catch prompt injection
- Security rules: never print secrets, confirm writes, prefer --dry-run
- JSON/table/yaml/csv output and --page-all NDJSON auto-pagination
- Browser OAuth or service-account authentication
Install This Skill
Add this skill to your favorite AI agent in a few steps.
Skill Content
Usage Instructions
Learn how to use this skill with different AI agents.
Example Usage
gws drive files list --params '{"pageSize": 5}'
gws sheets +read --spreadsheet ID --range "Sheet1!A1:D10"
gws gmail +triage
Description
Google Workspace's official agent skill for gws, a single command-line tool covering Drive, Gmail, Calendar, Sheets, Docs, Chat and Admin. This entry is the shared reference the other Workspace skills depend on: authentication, global flags, output formatting and the security rules every service skill inherits.
What makes the CLI itself unusual. It builds its command surface dynamically from Google's Discovery Service rather than shipping a hand-maintained command list, so endpoints Google adds appear without waiting for a CLI release. The repository ships more than 100 SKILL.md files — one per supported API, plus higher-level helpers such as gws gmail +triage, +reply (which handles threading itself), +forward and +watch, which streams new mail as NDJSON.
The part that matters for agent safety. --sanitize screens API responses through Google Cloud Model Armor to catch prompt injection before the content reaches the agent — a real concern when an agent is reading arbitrary inbound email or shared documents. The skill also encodes standing rules: never print secrets, always confirm before write or delete operations, and prefer --dry-run for anything destructive.
Practical detail it gets right. Output defaults to JSON with table, yaml and csv alternatives; --page-all auto-paginates into NDJSON with configurable page limits and delays; --params and --json carry query parameters and request bodies. It also warns about two shell traps that bite in practice — zsh expanding the ! in a Sheets range like Sheet1!A1:D10 as history expansion, and quoting rules for JSON arguments.
Installation. Pre-built binaries from GitHub Releases, npm install -g @googleworkspace/cli, Homebrew, Cargo or a Nix flake. Authenticate interactively with gws auth login or non-interactively via a service account through GOOGLE_APPLICATION_CREDENTIALS. Apache-2.0 licensed.
Related Skills
Official WordPress skill for Gutenberg block work: block.json, attributes and serialization, dynamic rendering, and the deprecation path that keeps existing content valid.
Persistent cross-session memory for coding agents: hooks capture each session, a local SQLite + vector store compresses it, and a mem-search skill reads it back.
The master skill in PlanetScale's official pack: runs a full read-only best-practices assessment of a database or org, then produces one evidence-backed report.
