[ai.agentserver] add azure-ai-agentserver-agentframework and azure-ai-agentserver-langgraph to agentserver

[lusu-msft]

Description

Please add an informative description that covers that changes made by the pull request and link all relevant issues.

If an SDK is being regenerated based on a new API spec, a link to the pull request containing these API spec changes should be included above.

All SDK Contribution checklist:

• The pull request does not introduce [breaking changes]
• CHANGELOG is updated for new features, bug fixes or other significant changes.
• I have read the contribution guidelines.

General Guidelines and Best Practices

• Title of the pull request is clear and informative.
• There are a small number of commits, each of which have an informative message. This means that previously merged commits do not appear in the history of the PR. For more information on cleaning up the commits in your PR, see this page.

Testing Guidelines

• Pull request includes test coverage for the included changes.

[Copilot]

Pull Request Overview

This PR introduces a comprehensive agent server framework for Azure AI, consisting of two main packages:

• azure-ai-agentserver-core: Core server infrastructure for hosting agents
• azure-ai-agentserver-langgraph: LangGraph-specific adapter implementation

The implementation provides a standardized way to host AI agents as HTTP services with OpenTelemetry observability, supporting both streaming and non-streaming responses.

Reviewed Changes

Copilot reviewed 158 out of 166 changed files in this pull request and generated 8 comments.

Show a summary per file

File	Description
shared_requirements.txt	Adds dependencies for agent server packages
sdk/agentserver/ci.yml, tests.yml, platform-matrix.json	CI/CD pipeline configuration for the new agentserver SDK
azure-ai-agentserver-core/*	Core package with base agent server, request handling, and ID generation
azure-ai-agentserver-langgraph/*	LangGraph adapter with state converters and streaming support
samples/*	Multiple sample implementations demonstrating various agent patterns

Comments suppressed due to low confidence (1)

sdk/agentserver/azure-ai-agentserver-langgraph/samples/custom_state/main.py:1

• Using eval() on user input is a security risk as it can execute arbitrary Python code. Consider using ast.literal_eval() for safe evaluation or a proper expression parser library.

[l0lawrence]

for the tests failing due to mindependency we can disable that as well until -core gets released, and we can do that in each pyproject.toml as well under the tools.build section -> and hopefully that makes this ci green ---> see pyporject.toml comment instead

[l0lawrence]

@lusu-msft fyi this pr is going to fail on bandit even though we are disabling it, bandit is a check that we actually require to run on all packages -> once we get this pr to a point where bandit is the only check failing that can be our equivalent to the pr being green and we can override it and merge. @scbedd will help unblock us on the eng sys side :)

[l0lawrence]

@lusu-msft can we add these two packages as artifacts to the ci.yml

azure-sdk-for-python/sdk/agentserver/ci.yml

Line 45 in 90a9226

- name: azure-ai-agentserver-core

, like how it is done for the packages under ai

azure-sdk-for-python/sdk/ai/ci.yml

Line 58 in 90a9226

- name: azure-ai-projects

[lmazuel]

Approving based on offline discussion about the status of the package. The Bandit job is related to this package not supporting 3.9, which has been approved. We manually checked using a more recent of Python that Bandit do not raise any security issue on the package.