feat: Add Node.js/TypeScript Semantic Kernel sample agent - #281
Open
Akshit-MSFT (Akshit0022) wants to merge 5 commits into
Open
feat: Add Node.js/TypeScript Semantic Kernel sample agent#281Akshit-MSFT (Akshit0022) wants to merge 5 commits into
Akshit-MSFT (Akshit0022) wants to merge 5 commits into
Conversation
Adds a new Node.js/TypeScript sample under nodejs/semantic-kernel/sample-agent/ demonstrating the Microsoft Agent 365 SDK with OpenAI function calling (Semantic Kernel-style patterns). Features: - Dual LLM support (Azure OpenAI / OpenAI) - MCP tool integration with @openai/agents SSE transport - Full A365 observability (InvokeAgent, Inference, ExecuteTool scopes) - Agentic authentication with token exchange - Notification handling (email, Word comments) - Conversation continuity with per-conversation chat history - Install/uninstall event handling - Express 5 server with CloudAdapter
Contributor
There was a problem hiding this comment.
Pull request overview
Adds a new Node.js/TypeScript “Semantic Kernel-style” sample agent under nodejs/semantic-kernel/sample-agent/, mirroring the existing .NET/Python semantic-kernel samples and the existing Node.js OpenAI/Claude samples. The sample demonstrates OpenAI Chat Completions tool-calling (local plugins + MCP tools), Agent 365 notifications, and A365 observability wiring.
Changes:
- Introduces a new Express-based agent host with
/api/messagesand JWT auth middleware. - Implements a Semantic Kernel-style function-calling loop (local plugins) plus MCP tool registration/execution.
- Adds full sample documentation, manifests, env template, and basic token caching for observability export.
Reviewed changes
Copilot reviewed 14 out of 16 changed files in this pull request and generated 8 comments.
Show a summary per file
| File | Description |
|---|---|
| nodejs/semantic-kernel/sample-agent/tsconfig.json | TypeScript build configuration for the sample |
| nodejs/semantic-kernel/sample-agent/package.json | Sample dependencies and scripts |
| nodejs/semantic-kernel/sample-agent/src/index.ts | Express server entry point and routing/auth middleware |
| nodejs/semantic-kernel/sample-agent/src/agent.ts | AgentApplication implementation (messages, notifications, install/uninstall, observability token preload) |
| nodejs/semantic-kernel/sample-agent/src/client.ts | Core “Semantic Kernel-style” client with tool loop + MCP integration + observability scope |
| nodejs/semantic-kernel/sample-agent/src/openai-config.ts | OpenAI/Azure OpenAI client selection and @openai/agents Azure configuration |
| nodejs/semantic-kernel/sample-agent/src/plugins.ts | Local plugin tools (terms & conditions accept/reject flow) |
| nodejs/semantic-kernel/sample-agent/src/token-cache.ts | In-memory token cache + custom observability token resolver |
| nodejs/semantic-kernel/sample-agent/ToolingManifest.json | MCP server manifest fallback for tooling discovery |
| nodejs/semantic-kernel/sample-agent/.env.template | Environment variable template for auth, tooling, and LLM config |
| nodejs/semantic-kernel/sample-agent/.gitignore | Ignores build output and secret/config files |
| nodejs/semantic-kernel/sample-agent/appManifest/manifest.json | Teams/M365 app manifest for the sample |
| nodejs/semantic-kernel/sample-agent/appManifest/color.png | Teams app icon (color) |
| nodejs/semantic-kernel/sample-agent/appManifest/outline.png | Teams app icon (outline) |
| nodejs/semantic-kernel/sample-agent/docs/design.md | Architecture/design notes for the sample |
| nodejs/semantic-kernel/sample-agent/README.md | End-to-end sample documentation (setup, run, troubleshooting) |
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Author
|
@microsoft-github-policy-service agree company="Microsoft" |
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Author
|
HI Grant Harris (@gwharris7), can you please review this and approve? |
Author
|
Hi Grant Harris (@gwharris7), kindly review this and take appropriate action. |
|
Grant Harris (@gwharris7) - Please review and provide your approvals. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Adds a new Node.js/TypeScript sample demonstrating how to build a production-ready Microsoft 365 agent using OpenAI function calling (Semantic Kernel-style patterns) with the Microsoft Agent 365 SDK.
This is the Node.js/TypeScript counterpart to the existing .NET Semantic Kernel sample (
dotnet/semantic-kernel/) and the Python Semantic Kernel sample (python/semantic-kernel/).What's included
New directory:
nodejs/semantic-kernel/sample-agent/src/index.ts/api/messagesendpoint and auth middlewaresrc/agent.tsAgentApplicationwith message routing, notifications, install/uninstall handlingsrc/client.tsSemanticKernelClientwith OpenAI Chat Completions function-calling loop, MCP tool integration, and A365 observability scopessrc/openai-config.tssrc/plugins.tssrc/token-cache.tsToolingManifest.json.env.templatepackage.jsontsconfig.jsonappManifest/manifest.jsonappManifest/color.pngappManifest/outline.pngdocs/design.mdREADME.md.gitignoreFeatures
AZURE_OPENAI_API_KEY/OPENAI_API_KEY)ToolingManifest.jsonfallback; uses@openai/agentsfor MCP server connections with SSE transportUSE_AGENTIC_AUTH=false(bearer token fallback)InvokeAgentScope→InferenceScope→ExecuteToolScopewithBaggageScopecontext propagation@microsoft/agents-a365-notificationsChatHistorymaintained across turnsCloudAdapter, compatible with Agents Playground, matching the pattern used by other Node.js samplesDependencies
openai^4.77.0@openai/agents^0.1.11express^5.1.0@microsoft/agents-hosting^1.2.2@microsoft/agents-activity^1.2.2@microsoft/agents-a365-observability^0.1.0-preview.125@microsoft/agents-a365-tooling-extensions-openai^0.1.0-preview.125@microsoft/agents-a365-notifications^0.1.0-preview.125@microsoft/agents-a365-runtime^0.1.0-preview.125Full list in
package.json.Testing
USE_AGENTIC_AUTH=falsewith bearer token)mcp_MailTools)gpt-4oRelated samples
This follows the same architecture and conventions as:
nodejs/claude/sample-agent/nodejs/openai/sample-agent/dotnet/semantic-kernel/sample-agent/(the .NET equivalent)python/semantic-kernel/sample-agent/(the Python equivalent)