编辑甄选 · OSS

GitHub 高星项目· 5 projects

由潜龙编辑团队甄选，每周更新最值得看的开源项目。我们看重的不是 Star 多少，而是它解决问题的方式。

全部主题高潜项目 ai ai-agent ai-agents ai-assistant ai-tools cheating claude-code claude-skills cluely cluely-alternative cluely-clone final-round-ai agent-framework artificial-intelligence automation developer-tools llm multi-agent open-source rust workflow-automation claude hacs home-assistant home-automation mcp model-context-protocol ai-hedge-fund ai-quant anthropic byok docker-compose financial-ai fintech investment-research llm-agent local-first c cli coding-agent coding-assistant llama-cpp local-llm openai openrouter repl agentic-coding ai-coding ai-governance context-engineering context-gateway context-intelligence copilot cursor csharp cuda cuda-programming deepseek dotnet dspark gemma gemma4 gguf glm inference agents analytics antigravity cli-tool codex cursor-ai insights kiro large-language-models agentic-workflows ai-workbench data-analysis deepseek-harness desktop-app dsh dsh-plugin electron batch-processing content-extraction data-collection firecrawl firecrawl-ai gemini-cli-extension javascript-rendering llm-tools mcp-server agent-client-protocol codex-cli course-downloader download-manager downloader hotmart-downloader instagram-downloader agent-skills ai-note-taking ai-second-brain claude-code-skill claude-memory claude-plugin claude-skill karpathy-llm-wiki agentic-ai ai-orchestration business-automation enterprise-ai erp fastapi global i18n pi pi-agent pi-desktop plugins ai-platform chatbot conversational-ai generative-ai llmops operating-system claude-code-skills agent-runtime api multi-model runtime ai-identity ai-memory digital-soul gemini-cli long-term-memory agent-memory memory rag copy-on-write embedded-database full-text-search hnsw hybrid-search langchain llamaindex mysql oceanbase python ai-sdk autonomous-agents durable-execution gemini human-in-the-loop a2a a2a-protocol agent-cli ai-agents-cli ai-workforce llm-runtime local-ai local-ai-agents agent-harness agent-observability agent-plugin agent-sandbox ai-infrastructure audit-logging devops docker amd gfx1151 gpu-inference hip inference-engine llm-serving long-context mixture-of-experts openai-api qwen acryl agent-context-relay agentic agentic-development-environment agentic-workflow agent-security ai-agent-permissions ai-safety api-gateway audit-trail authorization decionis execution-authority agent-workspace hermes-agent android apple-intelligence cpp diffusion-models edge flutter ios kotlin llamacpp multimodal ai-coding-assistant electron-app gui code-intelligence embeddings knowledge-base semantic-search agent agent-os grok-bot-alternative muse-alternative personal-ai self-hosted agent-loops agent-orchestration data-catalog data-governance data-integration data-lineage data-pipeline data-platform database-management er-diagram internal-tools cross-platform database-client database-gui database-tool mariadb postgresql ai-animation ai-video-generator capcut claude-agent-sdk content-creation ffmpeg image-to-video multi-provider novel-to-video knowledge-graph markdown muapi bm25 coding-agents git-native go golang llm-memory ai-code-review code-review devtools bun git-worktree bubbletea command-code grok agentops ecommerce java langgraph rabbitmq spring-boot github-actions harness ai-agent-skills codex-skills cross-runtime ai-cost-tracking gateway llm-observability llm-router ide novel groq linux macos nvidia parakeet speech-to-text openclaw ai-calling asterisk-ari inbound-calls no-code on-prem-voice-agent-platform open-source-voice-ai outbound-calls pipecat agent-improvement ai-recommendations continuous-improvement developer-productivity prompt-engineering reusable-skills claude-code-plugins marketplace browser web ai-coding-agent gentle-ai openspec chatops discord microsoft-teams canvas javascript nle reels video-editing video-editor zero-dependency anthropic-api chat inference-gateway lightseek responses-api routing ai-coding-workflow ai-development-dashboard ai-progress-tracking ai-workflow code-workflow coding-agent-orchestration cursor-agent development-kanban claude-ai local-echo opencode ai-observability agent-api conformance-testing harness-engineering hermes open-protocol acp

共 5 个项目·第 1 / 共 1

排序Star 最多最新发布

隐

OpenWhispr

隐私优先的跨平台语音听写工具

8.1k

OpenWhispr 是一个跨平台的语音转文本听写应用。简单说，它为用户提供了一个隐私优先的语音输入工具，既支持在本地运行 Nvidia Parakeet 或 Whisper 等模型，也允许用户自带密钥（BYOK）接入云端大模型。这个项目的方向挺直接：随着开源语音识别模型的成熟，很多人希望能在不泄...

JavaScript高潜项目990

让

SamurAIGPT

让大模型自己维护知识库

3.5k

别把它看成普通的 RAG 聊天框，它更像是一个自动化的知识整理工。简单说，你只需要把资料扔进去，Claude 或 Gemini 就会自己去阅读、提取知识，然后帮你维护一个持续更新、互相链接的 Wiki 系统。这里有个小看点，它不需要你配置复杂的 API 密钥，而是直接和 Claude Code、G...

Python高潜项目404

the-open-engine

Zeroshot - 执行与验证分离的代码变更编排工具

1.8k

在 AI 辅助编程的领域，Zeroshot 摸索出了一个挺有意思的架构模式：执行器（executor）与验证器（verifier）分离。简单说，它不是让一个大模型既当裁判又当运动员，而是通过独立的编排机制，让 AI 生成的代码变更必须经过独立的验证环节。这个项目是用 Rust 写的，主要以 CLI...

Rust高潜项目171

Deuz-AI

Deuz-SDK：零依赖的生产级智能体框架

697

这是一个面向生产环境的 TypeScript 智能体框架，最大的卖点是“零依赖”。它不仅提供了一个统一的流式 API 来调用 Claude、GPT、Gemini、DeepSeek 等主流模型，还内置了持久化执行、长期记忆、混合 RAG 以及 MCP 工具调用等高级特性。在构建复杂的 AI 智能体时...

TypeScript高潜项目1

smg-project

Rust 编写的高性能 LLM 网关

513

这个项目的方向挺直接，就是要做一个与底层推理引擎解耦的高性能大模型网关。SMG 用 Rust 编写，能够将 vLLM、TRT-LLM、SGLang 等各种推理后端的接口，统一转换为标准的 OpenAI 和 Anthropic API 格式。真正难的地方不是做个简单的 API 转发，而是如何在网关层...

Rust高潜项目159