Python: Add OrcaRouter as a named chat completion service

[kuswardhanietidims-svg]

Motivation and Context

Semantic Kernel is a model-agnostic SDK that empowers developers to build, orchestrate, and deploy AI agents and multi-agent systems, and one of its headline features is model flexibility — "connect to any LLM." Today the Python samples let you pick a provider by name from Services (OpenAI, Azure OpenAI, Anthropic, Mistral AI, Ollama, NVIDIA NIM, DeepSeek, ...), and each named provider is wired through a settings class with its own environment-variable prefix.

OrcaRouter is an OpenAI-compatible AI gateway built for both models and agents. Like OpenRouter, it exposes a provider/model namespace across many models — but it also combines adaptive routing, automatic failover, zero-markup inference, observability, guardrails, and agent-tool governance behind the same endpoint. Adding orcarouter as a first-class provider means this project's users can use that stack directly, without treating OrcaRouter as an anonymous custom base URL. It also runs gateway-level, zero-trust security for AI agents on the same endpoint — screening every prompt/response and governing every tool call on a default-deny basis, with no application code changes.

Description

This PR mirrors the existing DeepSeek entry (Services.DEEPSEEK → get_deepseek_chat_completion_service_and_request_settings) in python/samples/concepts/setup/chat_completion_services.py and adds an equivalent named OrcaRouter provider:

• Services.ORCAROUTER = "orcarouter" in the Services enum.
• OrcaRouterSettings(KernelBaseSettings) with the ORCAROUTER_ env prefix (ORCAROUTER_API_KEY, ORCAROUTER_CHAT_MODEL_ID).
• get_orcarouter_chat_completion_service_and_request_settings() which builds an OpenAIChatCompletion backed by an AsyncOpenAI client pointed at https://api.orcarouter.ai/v1, exactly like the DeepSeek wiring.
• Services.ORCAROUTER listed in the simple_chatbot.py sample's selectable-service comment.

Verified locally:

• ruff check and ruff format --check pass on both modified files.
• mypy reports no errors in the new code (the single existing mypy warning is a pre-existing Onnx type issue in chat_completion_services.py, unrelated to this change).
• L3 live test: with ORCAROUTER_API_KEY set and ORCAROUTER_CHAT_MODEL_ID=orcarouter/free, calling get_chat_completion_service_and_request_settings(Services.ORCAROUTER) and invoking the resulting service returned a successful chat completion.

Discord: discord.gg/YEubt8enRA · X: https://x.com/OrcaRouter

Contribution Checklist

• The code builds clean without any errors or warnings
• The PR follows the SK Contribution Guidelines and the pre-submission formatting script raises no violations
• All unit tests pass, and I have added new tests where possible
• I didn't break anyone 😄

---

I'm an engineer on the OrcaRouter team.

[Copilot]

Pull request overview

Adds OrcaRouter as a first-class, selectable chat completion provider in the Python samples, wired through the existing OpenAI-compatible connector pattern.

Changes:

• Introduces OrcaRouterSettings (ORCAROUTER_ env prefix) and Services.ORCAROUTER in chat_completion_services.py.
• Adds get_orcarouter_chat_completion_service_and_request_settings() that creates an OpenAIChatCompletion backed by an AsyncOpenAI client targeting https://api.orcarouter.ai/v1.
• Updates the simple_chatbot.py sample’s provider-selection comment to include Services.ORCAROUTER.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.

File	Description
python/samples/concepts/setup/chat_completion_services.py	Adds OrcaRouter settings, enum entry, service factory mapping, and OrcaRouter OpenAI-compatible service construction.
python/samples/concepts/chat_completion/simple_chatbot.py	Lists OrcaRouter as a selectable service in the sample’s comment.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[github-actions]

MAF Automated Review — Iteration 1

Result: No findings
Scope: full PR (1 commit(s)): 6e40cce2f62b
Model: claude-opus-4.8

Overview

This PR adds Services.ORCAROUTER as a selectable named chat completion service in
the Python concept samples, mirroring the existing DeepSeek entry and routing through
the OpenAI-compatible connector. The change is confined to two files under
python/samples/concepts/ with no library, public-API, or test surface affected. It
is well guarded: api_key/chat_model_id are checked before use so the SecretStr
is never dereferenced when None, the credential is unwrapped once and passed only to
the AsyncOpenAI client under TLS, and a dedicated ORCAROUTER_ env prefix avoids the
OPENAI_* collision the DeepSeek entry has. Residual items are Low-only: the
selectable-service comment was synced in one sample but not the ~13 others, and
ALL_SETTINGS.md was not updated (both consistent with existing repo state).

Reviewed the supplied pull-request change set across correctness, security/reliability, architecture, and failure behavior.
No publishable findings remained after source verification for this scope.

[kuswardhanietidims-svg]

@microsoft-github-policy-service agree