Senparc.AI.AgentKernel
0.1.14-preview5
Prefix Reserved
dotnet add package Senparc.AI.AgentKernel --version 0.1.14-preview5
NuGet\Install-Package Senparc.AI.AgentKernel -Version 0.1.14-preview5
<PackageReference Include="Senparc.AI.AgentKernel" Version="0.1.14-preview5" />
<PackageVersion Include="Senparc.AI.AgentKernel" Version="0.1.14-preview5" />
<PackageReference Include="Senparc.AI.AgentKernel" />
paket add Senparc.AI.AgentKernel --version 0.1.14-preview5
#r "nuget: Senparc.AI.AgentKernel, 0.1.14-preview5"
#:package Senparc.AI.AgentKernel@0.1.14-preview5
#addin nuget:?package=Senparc.AI.AgentKernel&version=0.1.14-preview5&prerelease
#tool nuget:?package=Senparc.AI.AgentKernel&version=0.1.14-preview5&prerelease
Senparc.AI AgentKernel module built on Microsoft Agent Framework, implementing Senparc.AI standard interfaces with support for Chat, Embedding, SpeechToText, TextToSpeech, TextToImage, vector storage, and multi-platform providers including OpenAI, Azure OpenAI, NeuCharAI, HuggingFace, FastAPI, Ollama, DeepSeek, Anthropic, Gemini, Qwen, Kimi, and XunFei
| Product | Versions Compatible and additional computed target framework versions. |
|---|---|
| .NET | net10.0 is compatible. net10.0-android was computed. net10.0-browser was computed. net10.0-ios was computed. net10.0-maccatalyst was computed. net10.0-macos was computed. net10.0-tvos was computed. net10.0-windows was computed. |
-
net10.0
- Azure.AI.OpenAI (>= 2.9.0-beta.1)
- Azure.AI.Projects (>= 2.1.0-beta.2)
- Azure.Identity (>= 1.21.0)
- Microsoft.Agents.AI (>= 1.8.0)
- Microsoft.Agents.AI.Foundry (>= 1.8.0-preview.260528.1)
- Microsoft.Agents.AI.Harness (>= 1.8.0-preview.260528.1)
- Microsoft.Agents.AI.Hosting.A2A.AspNetCore (>= 1.8.0-preview.260528.1)
- Microsoft.Agents.AI.OpenAI (>= 1.8.0)
- Microsoft.Agents.AI.Workflows (>= 1.8.0)
- Microsoft.Extensions.AI (>= 10.5.1)
- Microsoft.Extensions.VectorData.Abstractions (>= 10.1.0)
- Microsoft.SemanticKernel.Connectors.HuggingFace (>= 1.74.0-preview)
- Microsoft.SemanticKernel.Connectors.InMemory (>= 1.74.0-preview)
- Microsoft.SemanticKernel.Connectors.Ollama (>= 1.74.0-alpha)
- Microsoft.SemanticKernel.Connectors.Qdrant (>= 1.74.0-preview)
- Microsoft.SemanticKernel.Connectors.Redis (>= 1.74.1-preview)
- ModelContextProtocol (>= 1.0.0)
- OllamaSharp (>= 5.4.25)
- Senparc.AI (>= 0.27.4)
- Senparc.AI.AgentKernel.Providers.FastAPI (>= 0.1.0)
- Senparc.AI.AgentKernel.Providers.HuggingFace (>= 0.1.0)
NuGet packages (7)
Showing the top 5 NuGet packages that depend on Senparc.AI.AgentKernel:
| Package | Downloads |
|---|---|
|
Senparc.Xncf.XncfBuilder
Senparc.Xncf.XncfBuilder 快速生成 XNCF 模块基础程序代码 |
|
|
Senparc.Xncf.PromptRange
你的提示词(Prompt)训练场 |
|
|
Senparc.Xncf.AIKernel
AI 核心模块,基于 Senparc.AI 为所有 AI 项目提供基础能力 |
|
|
Senparc.Xncf.WeixinManager
NCF 模块:盛派官方发布的微信管理后台 使用此插件可以在 NCF 中快速集成微信公众号、小程序的部分基础管理功能,欢迎大家一起扩展! 微信 SDK 基于 Senparc.Weixin SDK 开发。 |
|
|
Senparc.Xncf.AgentsManager
Agents 管理模块 |
GitHub repositories
This package is not used by any popular GitHub repositories.
| Version | Downloads | Last Updated |
|---|---|---|
| 0.1.14-preview5 | 95 | 9/18/2026 |
| 0.1.14-preview3 | 48 | 9/16/2026 |
| 0.1.14-preview1 | 144 | 9/14/2026 |
| 0.1.12 | 406 | 8/13/2026 |
| 0.1.11 | 473 | 7/30/2026 |
| 0.1.10 | 323 | 7/15/2026 |
| 0.1.9.1 | 171 | 6/28/2026 |
| 0.1.9 | 183 | 6/28/2026 |
| 0.1.8.1 | 88 | 6/26/2026 |
| 0.1.8 | 78 | 6/24/2026 |
| 0.1.7 | 652 | 6/18/2026 |
| 0.1.5 | 91 | 6/16/2026 |
| 0.1.2.2 | 90 | 5/27/2026 |
| 0.1.2.1 | 85 | 5/26/2026 |
| 0.1.1.3 | 81 | 5/25/2026 |
| 0.1.1 | 81 | 5/25/2026 |
| 0.1.0 | 87 | 5/23/2026 |
| 0.1.0-preview.1 | 75 | 5/23/2026 |
[2026-05-21] v0.1.0-preview.1 Initial release based on Microsoft Agent Framework
1. AgentAiHandler / AgentKernelHelper: Chat, Embedding, multi-platform model integration
2. AiKernel / AIKernelBuilder fluent configuration API (KernelConfigExtensions)
3. AgentSession multi-turn conversations; BuildKernelAsync / BuildKernelWithAgentSessionAsync
4. EmbeddingDimensions, vector storage (InMemory / Qdrant / Redis, etc.)
5. Fix HttpMessageHandler chain, log streaming read, and HuggingFace extension naming
[2026-05-25] v0.1.1 Add TextSearchStore
[2026-05-26] v0.1.2 Open Streaming Chat mode for AgentKernel
[2026-05-30] v0.1.4 Add extension functions to, such as Arguments Replacement, Auto search function-calling target
[2026-05-30] v0.1.5 Add TextToImage functions
[2026-06-17] v0.1.6 Add SpeechToText / TextToSpeech interfaces, samples and tests
[2026-06-19] v0.1.7 Enhance SpeechToText / TextToSpeech support
1. AgentKernelHelper.ConfigSpeechToText / ConfigTextToSpeech: OpenAI, Azure OpenAI, and NeuCharAI
2. AiKernel.SpeechToTextAsync / TextToSpeechAsync: core speech APIs
3. KernelConfigExtensions: ConfigSpeechToTextModel / ConfigTextToSpeechModel / RunSpeechToTextAsync / RunTextToSpeechAsync
4. KernelBuilderExtension: AddOpenAIAudio / AddAzureOpenAIAudio / AddNeuCharAIAudio
5. Keep ConfigAudioToText / ConfigTextToAudio as legacy aliases
6. Add SttSample / TtsSample console samples and KernelConfigExtensionsSpeechTests
[2026-06-25] v0.1.8 Multi-platform Provider support expansion
1. ConfigChat / ConfigTextEmbeddingGeneration: HuggingFace, FastAPI, DeepSeek, Anthropic, Gemini, Qwen, Kimi, XunFei(Chat)
2. New provider packages: Senparc.AI.AgentKernel.Providers.HuggingFace / Providers.FastAPI (OpenAI-compatible IChatClient / IEmbeddingGenerator)
3. KernelBuilderExtension: AddOpenAICompatibleChatCompletion / AddOpenAICompatibleEmbedding for OpenAI-compatible platforms
4. HuggingFace default endpoint https://router.huggingface.co/v1; FastAPI supports custom OpenAI-compatible gateway
5. Anthropic Embedding throws at config time (official API does not provide Embedding)
6. Senparc.AI provider-specific Keys: AnthropicKeys, GeminiKeys, QwenKeys, KimiKeys, FastAPIKeys, HuggingFaceKeys
7. Add AgentKernelHelperPlatformSupportTests / ProviderFactoriesTests / SenparcAiSettingAdditionalPlatformTests
[2026-06-28] v0.1.9 Add MCP support
1. Add McpToolsetBuilder / McpToolsetResult / McpToolBindingMode / McpServerOption
2. Add McpSample console sample
[2026-06-28] v0.1.9.1 Fix: SenparcAiSetting.Endpoint
[2026-07-15] v0.1.10 Add: ChatOptionsSanitizer.SanitizeForModel
[2026-07-31] v0.1.11 Dependency update from Senparc.AI to 0.27.4
[2026-08-13] v0.1.12 A2A / IWantTo execution-path consistency
1. Inject AgentKernelHelper.HttpClient into Azure/OpenAI SDK client pipelines, so caller diagnostics, proxy and test handlers are actually used.
2. Add RunChatResponseAsync / RunChatStreamingAsync strict IWantToRun APIs with cancellation support and upstream exception propagation.
3. Preserve LastException in the legacy RunChatAsync compatibility result.
[2026-09-11] v0.1.13 Add first-class Microsoft Agent Framework Harness support
1. Add AgentKernelHarness and BuildHarnessAgentAsync so applications can use the native MAF HarnessAgent through AgentKernel.
2. Keep least-privilege defaults for file access, file memory, Skills scanning, and Hosted Web Search while retaining Todo, plan/execute, context compaction, and tool approval.
3. Keep the Harness integration isolated from the legacy Senparc.AI.Kernel package.
[2026-09-11] v0.1.14-preview1 Persist native Harness session state
1. Expose AgentKernelHarness.SerializeSessionAsync and restore serialized sessions through BuildHarnessAgentAsync.
[2026-09-17] v0.1.14-preview4 Require native Harness streaming
1. Keep the shared ChatOptions model capability sanitizer before native Harness construction.
2. Route every AgentKernelHarness.RunStreaming call directly to MAF RunStreamingAsync so providers cannot be silently downgraded to buffered responses.
[2026-09-17] v0.1.14-preview5 Normalize NeuChar model aliases before sampling-parameter filtering
1. Filter Temperature, TopP, PresencePenalty, and FrequencyPenalty for NeuChar-prefixed GPT-5 aliases such as NeuChar-gpt-5.6-luna.