agents-v2-py

Build container-based Foundry Agents using Azure AI Projects SDK with ImageBasedHostedAgentDefinition. Use when creating hosted agents that run custom code in Azure AI Foundry with your own container images. Triggers: "ImageBasedHostedAgentDefinition", "hosted agent", "container agent", "Foundry Agent", "create_version", "ProtocolVersionRecord", "AgentProtocol.RESPONSES", "custom agent image".

Install with Tessl CLI

npx tessl i github:haniakrim21/everything-claude-code --skill agents-v2-py

What are skills?

1.31x

Review — 77%

Does it follow best practices?

If you maintain this skill, you can automatically optimize it using the tessl CLI to improve its score:

npx tessl skill review --optimize ./path/to/skill

Learn more

Evaluation — 100%

↑ 1.31x

Agent success when using this skill

Validation — 10 / 11 Passed

Validation for skill structure

SKILL.md

Review

Evals

Evaluation results

100%

23%

Deploy a Custom Sentiment Analysis Agent to Azure AI Foundry

Core hosted agent creation with SDK

Criteria

Without context

With context

SDK version constraint

50%

100%

azure-identity in requirements

100%

DefaultAzureCredential import

100%

Correct model imports

50%

100%

create_version method used

100%

ImageBasedHostedAgentDefinition used

100%

ProtocolVersionRecord with RESPONSES

100%

Endpoint from environment

100%

No hardcoded secrets

100%

environment_variables dict

100%

Code interpreter tool format

100%

DefaultAzureCredential usage

100%

Without context: $0.5862 · 2m 34s · 29 turns · 27 in / 10,072 out tokens

With context: $0.4420 · 1m 18s · 25 turns · 63 in / 4,225 out tokens

100%

Production Agent with External Tool Integration

MCP tools, resource limits, production best practices

Criteria

Without context

With context

MCP tool type field

100%

MCP server_label field

100%

MCP server_url field

100%

file_search tool included

100%

CPU within valid range

100%

Memory within valid range

100%

No latest tag in production

100%

try/except around creation

100%

list_versions called

100%

delete_version called

100%

Endpoint from environment

100%

create_version method

100%

Without context: $0.6644 · 2m 37s · 32 turns · 30 in / 9,713 out tokens

With context: $0.4911 · 1m 22s · 25 turns · 291 in / 4,591 out tokens

100%

48%

Async Agent Provisioning Service

Async agent creation pattern

Criteria

Without context

With context

Async credential import

100%

Async client import

100%

Models imports correct

100%

Async credential context manager

100%

Async client context manager

100%

Await create_version

100%

ProtocolVersionRecord RESPONSES v1

100%

asyncio.run entrypoint

100%

Endpoint from environment

100%

SDK version constraint

100%

Without context: $0.4233 · 1m 52s · 19 turns · 25 in / 6,993 out tokens

With context: $0.3743 · 1m 2s · 21 turns · 27 in / 3,468 out tokens

Evaluated: 3 days ago
Agent: Claude Code
Model: Unknown

Table of Contents

Deploy a Custom Sentiment Analysis Agent to Azure AI Foundry Production Agent with External Tool Integration Async Agent Provisioning Service

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.