agents-v2-py

Build container-based Foundry Agents with Azure AI Projects SDK (ImageBasedHostedAgentDefinition). Use when creating hosted agents with custom container images in Azure AI Foundry.

2.56x

Quality

70%

Does it follow best practices?

Impact

100%

2.56x

Average score across 3 eval scenarios

Securityby

Advisory

Suggest reviewing before use

Optimize this skill with Tessl

npx tessl skill review --optimize ./skills/antigravity-agents-v2-py/SKILL.md

Evaluation results

100%

80%

Deploy a Document Processing Agent to Azure AI Foundry

Core hosted agent creation with Azure AI Projects SDK

Criteria

Without context

With context

SDK version pin

100%

DefaultAzureCredential auth

100%

Endpoint from env var

100%

Correct model imports

100%

create_version method

100%

ImageBasedHostedAgentDefinition

100%

Protocol version config

100%

CPU as string

100%

Memory Gi suffix

100%

Specific image tag

100%

Without context: $0.2863 · 2m 50s · 12 turns · 17 in / 5,046 out tokens

With context: $0.2518 · 2m 6s · 14 turns · 306 in / 2,579 out tokens

100%

50%

Research Assistant Agent with External Tool Integration

Multi-tool agent with MCP and environment configuration

Criteria

Without context

With context

Tools as list of dicts

100%

code_interpreter tool

100%

MCP tool type

100%

MCP server_label

50%

100%

MCP server_url

50%

100%

environment_variables dict

100%

Endpoint env var in container

100%

DefaultAzureCredential

100%

list_versions call

100%

Protocol version config

100%

Without context: $1.2703 · 10m 12s · 62 turns · 66 in / 11,412 out tokens

With context: $0.2518 · 45s · 16 turns · 54 in / 2,371 out tokens

100%

53%

Async Agent Deployment and Rollover Utility

Async agent lifecycle management with error handling

Criteria

Without context

With context

Async identity import

100%

Async client import

100%

Async credential context

100%

Async client context

100%

Await create_version

30%

100%

list_versions call

100%

delete_version params

100%

Try/except error handling

100%

Protocol version config

100%

Without context: $0.1994 · 2m 51s · 15 turns · 21 in / 3,232 out tokens

With context: $0.4423 · 3m 35s · 26 turns · 28 in / 4,525 out tokens

Repository: boisenoise/skills-collections
Commit: 5c5ae21

Evaluated: 17 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Deploy a Document Processing Agent to Azure AI Foundry Research Assistant Agent with External Tool Integration Async Agent Deployment and Rollover Utility

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.