CtrlK
BlogDocsLog inGet started
Tessl Logo

skill-creator

Guide for creating effective skills for AI coding agents working with Azure SDKs and Microsoft Foundry services. Use when creating new skills or updating existing skills.

66

2.68x
Quality

51%

Does it follow best practices?

Impact

94%

2.68x

Average score across 3 eval scenarios

SecuritybySnyk

Advisory

Suggest reviewing before use

Optimize this skill with Tessl

npx tessl skill review --optimize ./.github/skills/skill-creator/SKILL.md
SKILL.md
Quality
Evals
Security

Evaluation results

83%

70%

New Azure Tables SDK Skill

Azure SDK skill creation

Criteria
Without context
With context

Correct skill location

0%

100%

Skill name format

0%

100%

Description has triggers

0%

100%

No when-to-use in body

0%

100%

Installation before Authentication

0%

100%

DefaultAzureCredential auth

50%

100%

Cleanup in examples

50%

100%

No meta-docs in skill dir

0%

100%

Symlink created

0%

0%

Symlink in data category

0%

28%

Symlink short name

0%

71%

Content size discipline

60%

100%

100%

15%

Write Acceptance Criteria for a Skill

Acceptance criteria authoring

Criteria
Without context
With context

Correct output path

100%

100%

Visual correct/incorrect markers

90%

100%

Import path — correct example

100%

100%

Import path — incorrect example

100%

100%

Auth — DefaultAzureCredential correct

100%

100%

Auth — hardcoded credentials incorrect

70%

100%

Context managers correct

100%

100%

Grouped into sections

100%

100%

Skill naming patterns

100%

100%

Frontmatter patterns

80%

100%

Summary checklist

25%

100%

100%

93%

Create Test Scenarios for a Skill

Skill test scenario YAML authoring

Criteria
Without context
With context

Correct output path

0%

100%

Top-level config section

0%

100%

Config has model field

0%

100%

Config has max_tokens field

0%

100%

At least 3 scenarios

42%

100%

Each scenario has name

28%

100%

Each scenario has prompt

28%

100%

expected_patterns present

0%

100%

forbidden_patterns present

0%

100%

DefaultAzureCredential in expected

0%

100%

Hardcoded creds in forbidden

0%

100%

mock_response with complete code

0%

100%

Tags per scenario

0%

100%

Repository
microsoft/agent-skills
Evaluated
Agent
Claude Code
Model
Claude Sonnet 4.6

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.