gws-modelarmor
Google Model Armor: Filter user-generated content for safety.
57
Quality
47%
Does it follow best practices?
Impact
68%
1.47xAverage score across 3 eval scenarios
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/gws-modelarmor/SKILL.mdDiscovery
22%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This description is too brief and vague to effectively guide skill selection. While it names the specific Google product, it fails to describe concrete capabilities, lacks natural trigger terms users would say, and provides no explicit 'when to use' guidance. The description would benefit significantly from expansion.
Suggestions
Add a 'Use when...' clause with explicit triggers like 'Use when the user needs to check content for harmful material, moderate user submissions, or integrate Google's content safety API'
Expand specific capabilities: what types of safety checks (toxicity, hate speech, violence, etc.), what actions (classify, score, block, filter), and what content types (text, images, prompts)
Include natural trigger terms users would say: 'content moderation', 'harmful content detection', 'safety check', 'toxicity filter', 'content policy enforcement'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description only mentions 'filter user-generated content for safety' which is vague. It doesn't specify what types of content, what safety categories, or what concrete actions are performed (e.g., detect harmful content, block requests, return safety scores). | 1 / 3 |
Completeness | The description only weakly addresses 'what' (filter content for safety) and completely lacks a 'when' clause. There's no explicit guidance like 'Use when...' to help Claude know when to select this skill. | 1 / 3 |
Trigger Term Quality | Contains 'Google Model Armor' (product name), 'filter', 'user-generated content', and 'safety' which are somewhat relevant. However, it's missing natural variations users might say like 'content moderation', 'harmful content', 'toxicity detection', 'content filtering API', or 'safety screening'. | 2 / 3 |
Distinctiveness Conflict Risk | 'Google Model Armor' is a specific product name which helps distinctiveness, but 'filter content for safety' is generic enough to potentially conflict with other content moderation or safety-related skills. | 2 / 3 |
Total | 6 / 12 Passed |
Implementation
72%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill excels at conciseness and progressive disclosure, providing a clean overview that appropriately delegates to helper commands. However, it lacks concrete executable examples showing actual Model Armor usage (sanitizing content), and the workflow for using the tool could be more explicit with validation steps.
Suggestions
Add a concrete example showing a complete sanitize-prompt or sanitize-response command with actual parameters
Include a brief workflow showing the typical sequence: create template -> sanitize prompt -> handle response/errors
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely lean content that assumes Claude's competence. No unnecessary explanations of what Model Armor is or how filtering works. Every line serves a purpose. | 3 / 3 |
Actionability | Provides concrete commands for discovery (--help, gws schema) but lacks executable examples of actual usage. The helper command references are useful but the main skill doesn't show a complete working example. | 2 / 3 |
Workflow Clarity | The discovery workflow (browse -> inspect -> build params) is implicit but not explicitly sequenced. For a skill involving content filtering, there's no validation or error handling guidance. | 2 / 3 |
Progressive Disclosure | Excellent structure with clear one-level-deep references to helper commands. Points to prerequisite shared skill and organizes specialized tasks into separate linked files. | 3 / 3 |
Total | 10 / 12 Passed |
Validation
72%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 8 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
metadata_version | 'metadata.version' is missing | Warning |
metadata_field | 'metadata' should map string keys to string values | Warning |
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 8 / 11 Passed | |
- Reviewed
Table of Contents
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.