azure-ai-contentsafety-py

Azure AI Content Safety SDK for Python. Use for detecting harmful content in text and images with multi-severity classification.

Quality

68%

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/antigravity-azure-ai-contentsafety-py/SKILL.md

The canonical home for this skill is azure-ai-contentsafety-py in sickn33/antigravity-awesome-skills

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 an efficient, actionable SDK reference with concrete code for every common task and well-organized sections. It falls short of top marks mainly due to repeated boilerplate across code blocks, undefined shared variables, and no external reference files for the larger reference material.

Suggestions

Reduce repeated import/client-construction boilerplate across code blocks, or factor a shared setup snippet, to improve token efficiency.

Make each code block self-contained (define 'endpoint'/'key' or reference the Authentication snippet explicitly) so examples are copy-paste runnable.

Consider moving the bulkier reference tables and Best Practices into a one-level-deep reference file to better apply progressive disclosure.

Dimension	Reasoning	Score
Conciseness	The body is mostly lean code and tables with no padding about concepts Claude already knows; it assumes competence. Not a 5 because import statements and client construction are repeated in nearly every code block, and 'Best Practices' is somewhat list-heavy.	4 / 5
Actionability	Provides concrete, mostly copy-paste-ready code for text analysis, image analysis (file and URL), blocklist management, and severity modes, covering common cases. Not a 5 because several blocks reference undefined 'endpoint' and 'key' variables, so they are not standalone runnable without the Authentication section.	4 / 5
Workflow Clarity	An implicit sequence (Installation → Environment Variables → Authentication → Analyze) is conveyed through section ordering, and the operations are largely read-only analysis not requiring validation checkpoints. Not a 5 because there is no explicit stepwise workflow or feedback loop, and blocklist writes have no verify step.	4 / 5
Progressive Disclosure	Content is well-organized into clearly headed sections (Installation, Authentication, Analyze Text/Image, Blocklists, reference tables) with no nested references. Not a 5 because at ~200 lines everything is inlined in SKILL.md with no one-level-deep references for the bulkier reference material (e.g., full API/best-practices detail).	4 / 5
	Total	16 / 20 Passed

Description

61%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 clearly names the SDK and its core capability but relies on 'Use for...' rather than an explicit 'Use when...' trigger, capping completeness. It is specific and distinctive but could add natural synonyms and concrete trigger conditions.

Suggestions

Replace 'Use for...' with an explicit 'Use when...' clause naming concrete trigger phrases (e.g., 'Use when moderating user- or AI-generated content, filtering toxic or harmful text and images, or classifying content severity').

Add natural synonyms users say — 'content moderation', 'toxic content', 'profanity filtering' — to broaden trigger term coverage.

Mention blocklists and severity thresholds in the description so the capability list is more comprehensive.

Dimension	Reasoning	Score
Specificity	Names the domain ('Azure AI Content Safety SDK for Python') and 1-2 concrete actions ('detecting harmful content in text and images', 'multi-severity classification'), but coverage is not comprehensive — blocklists, thresholds, and 8-severity mode are absent. Fits the anchor 'Names domain and 1-2 concrete actions, but not comprehensive'; not a 4 because it does not list several specific actions.	3 / 5
Completeness	Has a clear 'what' (detect harmful content in text/images with multi-severity classification) but the 'when' is only weakly implied via 'Use for...' rather than an explicit 'Use when...' trigger clause. Per the guideline, a missing explicit 'Use when...' trigger caps completeness at 3.	3 / 5
Trigger Term Quality	Includes natural terms a user would say ('content safety', 'harmful content', 'detect', 'text and images'), giving good keyword coverage. Not a 5 because common synonyms like 'content moderation', 'toxic content', or 'profanity' are missing.	4 / 5
Distinctiveness Conflict Risk	'Azure AI Content Safety SDK for Python' is a clear, specific niche with distinct triggers, giving minimal conflict risk with unrelated skills. Not a 5 because 'harmful content in text and images' could still overlap with generic content-moderation skills.	4 / 5
	Total	14 / 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: boisenoise/skills-collections
Path: skills/antigravity-azure-ai-contentsafety-py/SKILL.md
Commit: 19c310e

Reviewed: about 17 hours 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.