Name: bootstrap-google-tools
Rating: 68 (1 reviews)
Author: google

bootstrap-google-tools

Install/auth CLIs the sandbox lacks - `gws` (Drive, Gmail, Sheets, Calendar), `gcloud`, `agents-cli`, `mcp-cli` (MCP servers). Use on "command not found" or before GCP/Workspace/MCP work.

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

75%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A dense, highly executable install/auth recipe with strong copy-paste commands, a clear check-then-verify pattern, and a useful failure cheat-sheet. Its main weakness is verbosity in the auth/security prose, which could be trimmed without losing the non-obvious sandbox-specific value.

Suggestions

Tighten the gws auth and GCP security prose: consolidate repeated 'don't/never' caveats into a single bullet list so the same warning isn't restated across sections.

Extract the detailed loopback-bridge OAuth recipe and the GCP security analysis into reference files linked from the main body to reduce inline length and improve progressive disclosure.

Add a short top-of-skill 'quick path' summary (check -> install missing -> verify) so the per-tool sections read as elaborations of one explicit master workflow.

Dimension	Reasoning	Score
Conciseness	Mostly high-value, sandbox-specific knowledge Claude would not already know, but the prose around gws auth and the GCP security section carries repeated emphatic caveats and explanatory padding that could be tightened, fitting the 'mostly efficient but some unnecessary explanation' anchor.	3 / 5
Actionability	Copy-paste-ready bash commands with exact env vars, install URLs, a complete mcp_servers.json example, and explicit verify steps after each install; placeholders (v22.x.x) are flagged for replacement, matching the fully-executable top anchor.	5 / 5
Workflow Clarity	Clear per-tool flow of check-first (command -v) -> install only what's missing -> verify, plus an if/then token-probe decision branch and a failure cheat-sheet for recovery; not a 5 because the menu structure lacks one unified master sequence with explicit validation checkpoints.	4 / 5
Progressive Disclosure	No bundle files are present, but the single SKILL.md is well-sectioned per tool with tables and a cheat-sheet, and it appropriately defers deeper usage to external skills (google-workspace, find-skills); the security and detailed OAuth-bridge material could arguably be split out, keeping it at 4 rather than 5.	4 / 5
	Total	16 / 20 Passed

Description

87%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A tight, third-person description that cleanly states what it does and when to reach for it, with concrete CLI names and a natural error-trigger phrase. Its only gap is modest action-verb variety (install/auth), which keeps specificity and trigger breadth at 4 rather than 5.

Dimension	Reasoning	Score
Specificity	Names two concrete actions ("Install/auth") applied to a comprehensive, explicitly enumerated set of CLIs (gws with its sub-surfaces, gcloud, agents-cli, mcp-cli), matching the 'several specific actions; minor gaps' anchor rather than the fully comprehensive 5-level action list.	4 / 5
Completeness	Explicitly answers both 'what' ("Install/auth CLIs the sandbox lacks") and 'when' ("Use on 'command not found' or before GCP/Workspace/MCP work") with concrete trigger phrases, matching the top anchor.	5 / 5
Trigger Term Quality	Includes the natural user-facing phrase "command not found" plus domain triggers "GCP/Workspace/MCP work" and the exact CLI names users would cite; good coverage but not exhaustive synonym-level breadth, so it sits at 4 rather than 5.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (bootstrap/auth sandbox-missing CLIs) with distinct triggers ("command not found", named CLIs) and minimal overlap with the per-surface usage skills it hands off to.	5 / 5
	Total	18 / 20 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: google/adk-samples
Path: core/python/long-horizon-harness/horizon/builtin_skills/bootstrap-google-tools/SKILL.md
Commit: 17f314a

Reviewed: 1 day ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.