CtrlK
BlogDocsLog inGet started
Tessl Logo

azure-ai-openai-dotnet

Azure OpenAI SDK for .NET. Client library for Azure OpenAI and OpenAI services. Use for chat completions, embeddings, image generation, audio transcription, and assistants.

Install with Tessl CLI

npx tessl i github:boisenoise/skills-collections --skill azure-ai-openai-dotnet
What are skills?

80

1.43x

Quality

73%

Does it follow best practices?

Impact

92%

1.43x

Average score across 3 eval scenarios

Optimize this skill with Tessl

npx tessl skill review --optimize ./skills/antigravity-azure-ai-openai-dotnet/SKILL.md
SKILL.md
Review
Evals

Evaluation results

80%

24%

Algorithm Analysis Assistant

Reasoning model integration

Criteria
Without context
With context

AzureOpenAIClient top-level

100%

100%

GetChatClient usage

100%

100%

DeveloperChatMessage for system prompt

0%

100%

ReasoningEffortLevel set

0%

0%

High MaxOutputTokenCount

100%

100%

Entra ID or credential auth

70%

100%

Environment variable usage

100%

100%

Output to file

100%

100%

Response text access

100%

100%

Without context: $0.2439 · 2m 41s · 14 turns · 17 in / 4,172 out tokens

With context: $0.3304 · 2m 12s · 20 turns · 55 in / 3,585 out tokens

96%

6%

Resume Data Extraction Pipeline

Structured JSON output extraction

Criteria
Without context
With context

ChatResponseFormat.CreateJsonSchemaFormat

100%

100%

jsonSchemaIsStrict: true

100%

100%

BinaryData.FromBytes for schema

100%

100%

JsonDocument parsing

50%

60%

Token usage logged

100%

100%

AzureOpenAIClient hierarchy

100%

100%

Credential from environment

100%

100%

DefaultAzureCredential or AzureKeyCredential

50%

100%

Schema additionalProperties false

100%

100%

Output to file

100%

100%

Without context: $0.5007 · 5m 35s · 28 turns · 36 in / 6,669 out tokens

With context: $0.4758 · 5m 23s · 26 turns · 327 in / 4,724 out tokens

100%

54%

Inventory Query Chatbot

Function calling with error handling

Criteria
Without context
With context

ChatTool.CreateFunctionTool

0%

100%

BinaryData.FromString for parameters

0%

100%

FinishReason ToolCalls check

33%

100%

ToolCalls iteration

50%

100%

Argument validation

90%

100%

Rate limit handling

50%

100%

RequestFailedException catch

0%

100%

Client reuse

100%

100%

DefaultAzureCredential or env var credentials

80%

100%

Output to file

100%

100%

Without context: $0.7234 · 6m 17s · 24 turns · 33 in / 13,168 out tokens

With context: $0.5625 · 5m 7s · 27 turns · 556 in / 7,810 out tokens

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.