Name: google-vertex-ai
Rating: 57.599999999999994 (1 reviews)
Author: membranedev

Blog Docs Log in Get started

google-vertex-ai

Google Vertex AI integration. Manage Projects. Use when the user wants to interact with Google Vertex AI data.

Quality

66%

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

Fix and improve this skill with Tessl

tessl review fix ./skills/google-vertex-ai/SKILL.md

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.

The body is highly actionable with concrete, executable Membrane CLI commands and a well-validated connection workflow, supported by clear tables and section structure. Its main weakness is a padded introductory paragraph and generic entity overview that restate knowledge Claude already has.

Suggestions

Delete or drastically shorten the opening 'Google Vertex AI is a machine learning platform...' paragraph and the generic entity Overview list, which restate concepts Claude already knows.

Fix the broken step references: label a 'Step 2' that 'skip to Step 2' points to, and reconcile the '1b' subheading with a missing '1a'.

Consider moving the Popular-actions and proxy-flags tables into a reference file to keep SKILL.md a lean overview, given the body exceeds 150 lines.

Dimension	Reasoning	Score
Conciseness	The opening paragraph ('Google Vertex AI is a machine learning platform that allows data scientists...') and the generic entity Overview explain concepts Claude already knows, but the bulk of the body is lean commands and tables.	3 / 5
Actionability	Copy-paste-ready commands throughout (install, login, headless login, connection ensure, poll, action list/run, request proxy) with real flags, plus concrete Popular-actions and proxy-flags tables covering common cases.	5 / 5
Workflow Clarity	Clear sequence from install to running actions with strong validation on the connection flow (state machine plus polling/retry), but 'skip to Step 2' has no labeled Step 2 and a '1b' appears with no '1a'.	4 / 5
Progressive Disclosure	Self-contained ~157-line doc with clear section headers and well-organized tables and no nested references; over the 50-line simple-skill threshold so not a clean 5, with minor organization gaps in step numbering.	4 / 5
	Total	16 / 20 Passed

Description

57%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.

The description answers both what and when with an explicit 'Use when' clause and a distinctive product name, but its single action ('Manage Projects') is generic and does not reflect the skill's actual ML capabilities. Trigger coverage is limited to the product name with no task-level synonyms.

Suggestions

Replace 'Manage Projects' with concrete Vertex AI actions, e.g. 'Tune Gemini models, manage endpoints, generate content, and create embeddings.'

Broaden the trigger with task-oriented phrases: 'Use when the user wants to tune or fine-tune Gemini models, manage Vertex AI endpoints, generate content, or create embeddings.'

Keep the third-person voice but make 'when' specific to ML tasks rather than the generic 'interact with Google Vertex AI data'.

Dimension	Reasoning	Score
Specificity	Names the domain ('Google Vertex AI integration') but the only action cited is 'Manage Projects', which is generic, oddly capitalized, and misrepresents a skill whose body covers tuning jobs, models, endpoints, and embeddings.	2 / 5
Completeness	Both 'what' ('integration. Manage Projects.') and an explicit 'when' ('Use when the user wants to interact with Google Vertex AI data.') are present, but the 'when' is generic rather than concrete trigger phrases.	4 / 5
Trigger Term Quality	The trigger relies on the product name 'Google Vertex AI' / 'Vertex AI data', a natural keyword, but omits task-oriented synonyms (tuning, models, endpoints, embeddings) users would actually say.	3 / 5
Distinctiveness Conflict Risk	It names a specific product giving it a clear niche, but 'Manage Projects' and 'interact with data' leave minor overlap risk with adjacent Google Cloud / AI skills.	4 / 5
	Total	13 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: membranedev/application-skills
Path: skills/google-vertex-ai/SKILL.md
Commit: f484c82

Reviewed: 5 days 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.