diff --git a/README.md b/README.md index 56493ea..45f52a0 100644 --- a/README.md +++ b/README.md @@ -18,6 +18,11 @@ English/[简体中文](https://github.com/deepseek-ai/awesome-deepseek-integrati ### Applications + + + + + @@ -93,7 +98,7 @@ English/[简体中文](https://github.com/deepseek-ai/awesome-deepseek-integrati - + @@ -101,6 +106,23 @@ English/[简体中文](https://github.com/deepseek-ai/awesome-deepseek-integrati + + + + + @@ -111,8 +133,8 @@ English/[简体中文](https://github.com/deepseek-ai/awesome-deepseek-integrati - - + + @@ -183,12 +205,21 @@ English/[简体中文](https://github.com/deepseek-ai/awesome-deepseek-integrati - + + + + +
Icon Quantalogic QuantaLogic is a ReAct (Reasoning & Action) framework for building advanced AI agents.
Icon Chatbox Raycast Raycast is a productivity tool for macOS that lets you control your tools with a few keystrokes. It supports various extensions including DeepSeek AI.
Icon Nice Prompt Nice Prompt Organize, share and use your prompts in your code editor, with Cursor and VSCode。
Icon Nice Prompt Nice Prompt Organize, share and use your prompts in your code editor, with Cursor and VSCode。
PHP Client Deepseek PHP Client is a robust and community-driven PHP client library for seamless integration with the Deepseek API.
+ DeepSwiftSeek Logo + + DeepSwiftSeek + + DeepSwiftSeek is a lightweight yet powerful Swift client library, pretty good integration with the DeepSeek API. + It provides easy-to-use Swift concurrency for chat, streaming, FIM (Fill-in-the-Middle) completions, and more. +
Laravel Integration Laravel Integration Laravel wrapper for Deepseek PHP client, to seamless deepseek API integration with laravel applications. Zotero is a free, easy-to-use tool to help you collect, organize, annotate, cite, and share research.
Icon SiYuan Icon SiYuan SiYuan is a privacy-first personal knowledge management system that supports complete offline usage, as well as end-to-end encrypted data sync.
16x Prompt 16x Prompt is an AI coding tool with context management. It helps developers manage source code context and craft prompts for complex coding tasks on existing codebases.
Icon PeterCat A conversational Q&A agent configuration system, self-hosted deployment solutions, and a convenient all-in-one application SDK, allowing you to create intelligent Q&A bots for your GitHub repositories.
### AI Agent frameworks + + + + + @@ -214,10 +245,15 @@ English/[简体中文](https://github.com/deepseek-ai/awesome-deepseek-integrati - + + + + + +
Icon smolagents The simplest way to build great agents. Agents write python code to call tools and orchestrate other agents. Priority support for open models like DeepSeek-R1!
Icon YoMoAn autonomous AI agent on ICP, leveraging LLMs like DeepSeek for on-chain decision-making. Alice combines real-time data analysis with a playful personality to manage tokens, mine BOB, and govern ecosystems.
Icon Icon Upsonic Upsonic offers a cutting-edge enterprise-ready agent framework where you can orchestrate LLM calls, agents, and computer use to complete tasks cost-effectively.
Icon ATTPs A foundational protocol framework for trusted communication between agents. Any agents based on DeepSeek, By integrating with the ATTPs SDK, can access features such as agent registration, sending verifiable data, and retrieving verifiable data. So that it can make trusted communication with agents from other platforms.
### RAG frameworks @@ -228,6 +264,13 @@ English/[简体中文](https://github.com/deepseek-ai/awesome-deepseek-integrati RAGFlow An open-source RAG (Retrieval-Augmented Generation) engine based on deep document understanding. It offers a streamlined RAG workflow for businesses of any scale, combining LLM (Large Language Models) to provide truthful question-answering capabilities, backed by well-founded citations from various complex formatted data. + + Icon + DeepSearcher + DeepSearcher combines powerful LLMs (DeepSeek, OpenAI, etc.) and Vector Databases (Milvus, etc.) to perform search, evaluation, and reasoning based on private data, providing highly accurate answer and comprehensive report. + + + ### Solana frameworks @@ -235,23 +278,27 @@ English/[简体中文](https://github.com/deepseek-ai/awesome-deepseek-integrati - +
Icon Solana Agent Kit Solana Agent Kit An open-source toolkit for connecting AI agents to Solana protocols. Now, any agent, using any Deepseek LLM, can autonomously perform 60+ Solana actions:
- ### Synthetic data curation + + + + + +
Icon Curator An open-source tool to curate large scale datasets for post-training LLMs.
Icon Kiln Generate synthetic datasets and distill R1 models into custom fine-tunes.
- ### IM Application Plugins @@ -315,6 +362,11 @@ English/[简体中文](https://github.com/deepseek-ai/awesome-deepseek-integrati + + + + +
RssFlow An intelligent RSS reader browser extension with AI-powered RSS summarization and multi-dimensional feed views. Supports DeepSeek model configuration for enhanced content understanding.
Icon Typral Fast AI writer assistant - Let AI help you quickly improve article, paper, text...
### VS Code Extensions @@ -330,6 +382,31 @@ English/[简体中文](https://github.com/deepseek-ai/awesome-deepseek-integrati Cline Meet Cline, an AI assistant that can use your CLI aNd Editor. + + Icon + AI Commit + Use AI to generate git commit messages in VS Code. + + + +### Visual Studio Extensions + + + + + + + + + + + + + + + + +
Icon Comment2GPT Use OpenAI ChatGPT, Google Gemini, Anthropic Claude, DeepSeek and Ollama through your comments
Icon CodeLens2GPT Use OpenAI ChatGPT, Google Gemini, Anthropic Claude, DeepSeek and Ollama through the CodeLens
Icon Unity Code Assist Lite Code assistance for Unity scripts
### neovim Extensions @@ -350,6 +427,11 @@ English/[简体中文](https://github.com/deepseek-ai/awesome-deepseek-integrati codecompanion.nvim AI-powered coding, seamlessly in Neovim. + + Icon + minuet-ai.nvim + Minuet offers code completion as-you-type from popular LLMs including Deepseek, OpenAI, Gemini, Claude, Ollama, Codestral, and more. + ### JetBrains Extensions @@ -392,13 +474,18 @@ English/[简体中文](https://github.com/deepseek-ai/awesome-deepseek-integrati -### Cursor +### Native AI Code Editor - + + + + + +
Icon Cursor ‍The AI Code Editor‍The AI Code Editor based on VS Code
Icon WindSurf Another AI Code Editor based on VS Code by Codeium
@@ -417,9 +504,24 @@ English/[简体中文](https://github.com/deepseek-ai/awesome-deepseek-integrati +### Security + + + + + + + +
Icon CodeGate CodeGate: secure AI code generation
+ ### Others + + + + + @@ -490,5 +592,4 @@ English/[简体中文](https://github.com/deepseek-ai/awesome-deepseek-integrati -
🐠 Abso TypeScript SDK to interact with any LLM provider using the OpenAI format.
Icon ShellOracle DeepSeek Alfred WorkFlow DeekSeek integrations in Alfred.
diff --git a/README_cn.md b/README_cn.md index ad8ce53..328cf12 100644 --- a/README_cn.md +++ b/README_cn.md @@ -18,6 +18,11 @@ ### 应用程序 + + + + + @@ -89,8 +94,8 @@ - - + + @@ -141,6 +146,11 @@ + + + + +
Icon Quantalogic QuantaLogic 是一个 ReAct（推理和行动）框架，用于构建高级 AI 代理。
Icon Chatbox Zotero 是一款免费且易于使用的文献管理工具，旨在帮助您收集、整理、注释、引用和分享研究成果。
Icon 思源笔记 Icon 思源笔记 思源笔记是一款隐私优先的个人知识管理系统，支持完全离线使用，并提供端到端加密的数据同步功能。
Story-Flicks 通过一句话即可快速生成高清故事短视频，支持 DeepSeek 等模型。
Icon PeterCat 我们提供对话式答疑 Agent 配置系统、自托管部署方案和便捷的一体化应用 SDK，让您能够为自己的 GitHub 仓库一键创建智能答疑机器人，并快速集成到各类官网或项目中， 为社区提供更高效的技术支持生态。
### AI Agent 框架 @@ -161,6 +171,11 @@ Alice 一个基于 ICP 的自主 AI 代理，利用 DeepSeek 等大型语言模型进行链上决策。Alice 结合实时数据分析和独特的个性，管理代币、挖掘 BOB 并参与生态系统治理。 + + 图标 + ATTPs + 一个用于Agent之间可信通信的基础协议框架，基于DeekSeek的Agent，可以接入ATTPs的SDK，获得注册Agent，发送可验证数据，获取可验证数据等功能，从而与其他平台的Agent进行可信通信。 + ### RAG 框架 @@ -171,6 +186,11 @@ RAGFlow 一款基于深度文档理解构建的开源 RAG（Retrieval-Augmented Generation）引擎。RAGFlow 可以为各种规模的企业及个人提供一套精简的 RAG 工作流程，结合大语言模型（LLM）针对用户各类不同的复杂格式数据提供可靠的问答以及有理有据的引用。 + + Icon + DeepSearcher + DeepSearcher 结合强大的 LLM（DeepSeek、OpenAI 等）和向量数据库（Milvus 等），根据私有数据进行搜索、评估和推理，提供高度准确的答案和全面的报告。 + ### Solana 框架 @@ -246,6 +266,11 @@ RssFlow 一款智能的RSS阅读器浏览器扩展，具有AI驱动的RSS摘要和多维度订阅视图功能。支持配置DeepSeek模型以增强内容理解能力。 + + Icon + Typral + 超快的AI写作助手 - 让AI帮你快速优化日报,文章,文本等等... + ### VS Code 插件 @@ -261,6 +286,11 @@ Cline Cline 是一款能够使用您的 CLI 和编辑器的 AI 助手。 + + Icon + AI Commit + 使用 AI 生成 git commit message 的 VS Code 插件。 + ### neovim 插件 @@ -276,6 +306,11 @@ llm.nvim 免费的大语言模型插件，让你在Neovim中与大模型交互，支持任意一款大模型，比如Deepseek，GPT，GLM，kimi或者本地运行的大模型(比如ollama) + + Icon + minuet-ai.nvim + Minuet 提供实时代码补全功能，支持多个主流大语言模型，包括 Deepseek、OpenAI、Gemini、Claude、Ollama、Codestral 等。 + Icon codecompanion.nvim @@ -298,9 +333,29 @@ +### AI Code编辑器 + + + + + + + + + + + + +
Icon Cursor 基于VS Code进行扩展的AI Code编辑器
Icon WindSurf 另一个基于VS Code的AI Code编辑器，由Codeium出品
+ ### 其它 + + + + @@ -339,6 +394,5 @@ - - +

🐠

 Abso + TypeScript SDK 使用 OpenAI 格式与任何 LLM 提供商进行交互。
Icon ShellOracle Icon DeepSeek Alfred WorkFlow 在 Mac Alfred App 中快速调用 DeekSeek 的 workflow。
diff --git a/README_ja.md b/README_ja.md index 8784e05..44c6457 100644 --- a/README_ja.md +++ b/README_ja.md @@ -6,7 +6,7 @@ # Awesome DeepSeek Integrations ![Awesome](https://cdn.rawgit.com/sindresorhus/awesome/d7305f38d29fed78fa85652e3a63e154dd8e8829/media/badge.svg) -DeepSeek APIを人気のソフトウェアに統合します。APIキーを取得するには、[DeepSeek Open Platform](https://platform.deepseek.com/)にアクセスしてください。 +DeepSeek API を人気のソフトウェアに統合します。API キーを取得するには、[DeepSeek Open Platform](https://platform.deepseek.com/)にアクセスしてください。 [English](https://github.com/deepseek-ai/awesome-deepseek-integration/blob/main/README.md)/[简体中文](https://github.com/deepseek-ai/awesome-deepseek-integration/blob/main/README_cn.md)/日本語 @@ -18,6 +18,11 @@ DeepSeek APIを人気のソフトウェアに統合します。APIキーを取 ### アプリケーション + + + + + @@ -99,8 +104,8 @@ DeepSeek APIを人気のソフトウェアに統合します。APIキーを取 - - + + @@ -138,9 +143,14 @@ DeepSeek APIを人気のソフトウェアに統合します。APIキーを取 + + + + +
Icon Quantalogic QuantaLogicは、高度なAIエージェントを構築するためのReAct（推論と行動）フレームワークです。
Icon Chatbox Zoteroは、研究成果を収集、整理、注釈、引用、共有するのに役立つ無料で使いやすいツールです。
Icon SiYuan Icon SiYuan SiYuanは、完全にオフラインで使用できるプライバシー優先の個人知識管理システムであり、エンドツーエンドの暗号化データ同期を提供します。
AgenticFlow AgenticFlowは、マーケターがAIエージェントのためのエージェンティックAIワークフローを構築するためのノーコードプラットフォームであり、数百の毎日のアプリをツールとして使用します。
Icon PeterCat 会話型Q&Aエージェントの構成システム、自ホスト型デプロイメントソリューション、および便利なオールインワンアプリケーションSDKを提供し、GitHubリポジトリのためのインテリジェントQ&Aボットをワンクリックで作成し、さまざまな公式ウェブサイトやプロジェクトに迅速に統合し、コミュニティのためのより効率的な技術サポートエコシステムを提供します。
-### AIエージェントフレームワーク +### AI エージェントフレームワーク @@ -148,9 +158,14 @@ DeepSeek APIを人気のソフトウェアに統合します。APIキーを取 + + + + +
Anda 高度にコンポーザブルで自律的かつ永続的な記憶を持つAIエージェントネットワークを構築するために設計されたRustフレームワーク。
Icon ATTPs エージェント間の信頼できる通信のための基本プロトコルフレームワークです。利用者はATTPsのSDKを導入することで、エージェントの登録、検証可能なデータの送信、検証可能なデータの取得などの機能を利用することができます。
-### RAGフレームワーク +### RAG フレームワーク @@ -158,9 +173,14 @@ DeepSeek APIを人気のソフトウェアに統合します。APIキーを取 + + + + +
RAGFlow 深い文書理解に基づいたオープンソースのRAG（Retrieval-Augmented Generation）エンジン。RAGFlowは、あらゆる規模の企業や個人に対して、ユーザーのさまざまな複雑な形式のデータに対して信頼性のある質問応答と根拠のある引用を提供するための簡素化されたRAGワークフローを提供します。
Icon DeepSearcher DeepSearcher は、強力な大規模言語モデル（DeepSeek、OpenAI など）とベクトルデータベース（Milvus など）を組み合わせて、私有データに基づく検索、評価、推論を行い、高精度な回答と包括的なレポートを提供します。
-### Solanaフレームワーク +### Solana フレームワーク @@ -170,7 +190,7 @@ DeepSeek APIを人気のソフトウェアに統合します。APIキーを取
-### IMアプリケーションプラグイン +### IM アプリケーションプラグイン @@ -228,9 +248,19 @@ DeepSeek APIを人気のソフトウェアに統合します。APIキーを取 + + + + + + + + + +
RssFlow AIを活用したRSS要約と多次元フィードビューを備えたインテリジェントなRSSリーダーブラウザ拡張機能。コンテンツ理解を強化するためのDeepSeekモデル設定をサポートしています。
Icon Ncurator ナレッジベース AI Q&Aアシスタント – AIがあなたの知識の整理と分析をお手伝いします
Icon Typral 超高速AIライティングアシスタント - AIがあなたの日報、記事、テキストなどを素早く最適化します
-### VS Code拡張機能 +### VS Code 拡張機能 @@ -243,9 +273,14 @@ DeepSeek APIを人気のソフトウェアに統合します。APIキーを取 + + + + +
Cline Clineは、CLIとエディタを使用できるAIアシスタントです。
Icon AI Commit VS Code で AI を使用して git commit message を生成するプラグイン。
-### neovim拡張機能 +### neovim 拡張機能 @@ -265,7 +300,7 @@ DeepSeek APIを人気のソフトウェアに統合します。APIキーを取
-### JetBrains拡張機能 +### JetBrains 拡張機能 @@ -295,7 +330,7 @@ DeepSeek APIを人気のソフトウェアに統合します。APIキーを取
-### Cursor +### AI コードエディタ @@ -303,6 +338,11 @@ DeepSeek APIを人気のソフトウェアに統合します。APIキーを取 + + + + +
Cursor ‍AIコードエディタ
Icon WindSurf CodeiumによるVS CodeをベースにしたのAIコードエディタ
### Emacs @@ -323,6 +363,11 @@ DeepSeek APIを人気のソフトウェアに統合します。APIキーを取 ### その他 + + + + + diff --git a/docs/ATTPs/README.md b/docs/ATTPs/README.md new file mode 100644 index 0000000..4b6360f --- /dev/null +++ b/docs/ATTPs/README.md @@ -0,0 +1,379 @@ + +# APRO-COM/ATTPs-framework + +Foundation framework that enables advanced agent based on DeepSeek interactions, data verification, and price queries with [ATTPs Protocol](https://docs.apro.com/attps) . It streamlines agent creation, verification processes, and provides a flexible framework for building robust agent-based solutions. + +For more details about ATTPs, you can see the [whitepaper here](https://www.apro.com/attps.pdf) + +## Overview + +The ATTPs framework bridges agent-based logic with the DeepSeek. It handles agent registration, data verification, and price queries, empowering both automated and user-driven workflows. + +## Features + +### Agent Operations +- **Agent Creation**: Deploy new agents with custom settings +- **Registration**: Register agents on-chain or via standardized processes +- **Multi-Signer Framework**: Supports threshold-based approval flows + +### Data Verification +- **Chain Validation**: Verify data authenticity on-chain +- **Transaction Execution**: Handle verification logic with built-in security checks +- **Auto-Hashing**: Convert raw data to hashed formats when needed +- **Metadata Parsing**: Validate content type, encoding, and compression + +### Price Queries +- **Live Price Data**: Fetch price information for various pairs +- **Format Validation**: Normalize user query inputs to standard trading-pair formats +- **APIs Integration**: Retrieve real-time or near-real-time pricing information + +## Security Features + +### Access Control +- **Private Key Management**: Safe usage of private keys for transaction signing +- **Environment Variables**: Secure injection of credentials +- **On-Chain Validation**: Leverage on-chain contract checks + +### Verification +- **Input Validation**: Strict schema checks before on-chain operations +- **Transaction Receipts**: Provide verifiable transaction details +- **Error Handling**: Detailed error logs for quick debugging + +## Performance Optimization + +1. **Cache Management** + - Implement caching for frequent queries + - Monitor retrieval times and cache hits + +2. **Network Efficiency** + - Batch requests where possible + - Validate response parsing to reduce overhead + +## System Requirements +- Node.js 16.x or higher +- Sufficient network access to on-chain endpoints +- Basic configuration of environment variables +- Minimum 4GB RAM recommended + +## Troubleshooting + +1. **Invalid Agent Settings** + - Ensure signers and threshold are correct + - Validate agentHeader for proper UUIDs and numeric values + +2. **Verification Failures** + - Check the input data formats + - Confirm environment variables are set + +3. **Price Query Errors** + - Verify the trading pair format + - Check external API availability + +## Safety & Security + +1. **Credential Management** + - Store private keys securely + - Do not commit secrets to version control + +2. **Transaction Limits** + - Configure thresholds to mitigate abuse + - Log transaction attempts and failures + +3. **Monitoring & Logging** + - Track unusual activity + - Maintain detailed audit logs + + +# Usage with js + +## Installation + +```bash +npm install ai-agent-sdk-js +``` + +## Configuration + +Configure the plugin by setting environment variables or runtime settings: +- APRO_RPC_URL +- APRO_PROXY_ADDRESS +- APRO_PRIVATE_KEY +- APRO_CONVERTER_ADDRESS +- APRO_AUTO_HASH_DATA + +## Usage with js sdk + +To use the AI Agent SDK, import the library and create an instance of the `Agent` class: + +```typescript +import { AgentSDK } from 'ai-agent-sdk-js' + +const agent = new AgentSDK({ + rpcUrl: 'https://bsc-testnet-rpc.publicnode.com', + privateKey: '', + proxyAddress: '', +}) + +// if you want the SDK to hash the data automatically +const autoHashAgent = new AgentSDK({ + rpcUrl: 'https://bsc-testnet-rpc.publicnode.com', + privateKey: '', + proxyAddress: '', + autoHashData: true, + converterAddress: '', +}) +``` + +To create a new agent, call the `createAndRegisterAgent` method: + +```typescript +import type { AgentSettings, TransactionOptions } from 'ai-agent-sdk-js' +import { randomUUID } from 'node:crypto' +import { parseUnits } from 'ethers' + +// prepare the agent settings +const agentSettings: AgentSettings = { + signers: [], + threshold: 3, + converterAddress: '', + agentHeader: { + messageId: randomUUID(), + sourceAgentId: randomUUID(), + sourceAgentName: 'AI Agent SDK JS', + targetAgentId: '', + timestamp: Math.floor(Date.now() / 1000), + messageType: 0, + priority: 1, + ttl: 3600, + }, +} + +// prepare the transaction options +const nonce = await agent.getNextNonce() +const transactionOptions: TransactionOptions = { + nonce, + gasPrice: parseUnits('1', 'gwei'), + gasLimit: BigInt(2000000), +} + +const tx = await agent.createAndRegisterAgent({ agentSettings, transactionOptions }) + +// or you can leave the transaction options empty, the SDK will use the auto-generated values +// const tx = await agent.createAndRegisterAgent({ agentSettings }) +``` + +The SDK also provides the tool to extract the new agent address from the transaction receipt: + +```typescript +import { parseNewAgentAddress } from 'ai-agent-sdk-js' + +const receipt = await tx.wait() +const agentAddress = parseNewAgentAddress(receipt) +``` + +To verify the data integrity, call the `verify` method: + +```typescript +import type { MessagePayload } from 'ai-agent-sdk-js' +import { hexlify, keccak256, toUtf8Bytes } from 'ethers' + +// prepare the payload +const data = hexlify(toUtf8Bytes('Hello World!')) +const dataHash = keccak256(data) +const payload: MessagePayload = { + data, + dataHash, + signatures: [ + { + r: '', + s: '', + v: 1, // 1, 0, 27, 28 are allowed + }, + // ... + ], + metadata: { + contentType: '', + encoding: '', + compression: '', + }, +} + +const tx = await agent.verify({ payload, agent: '', digest: '' }) +``` + +If the data is obtained from the APRO DATA pull service, you can use the auto-hash feature: + +```typescript +import type { MessagePayload } from 'ai-agent-sdk-js' + +const payload: MessagePayload = { + data: '0x...', + signatures: [ + { + r: '', + s: '', + v: 1, // 1, 0, 27, 28 are allowed + }, + // ... + ], + metadata: { + contentType: '', + encoding: '', + compression: '', + }, +} + +// When +const tx = await autoHashAgent.verify({ payload, agent: '', digest: '' }) +``` + +For more examples, see the [test](https://github.com/APRO-com/ai-agent-sdk-js/tree/main/test) cases. + + + +# Usage with Python + +## Installation + +```bash +$ pip3 install ai-agent-sdk + +``` + +## Usage with Python SDK + +### Initialize AgentSDK + +```python +from ai_agent.agent import AgentSDK + +AGENT_PROXY_ADDRESS = "0x07771A3026E60776deC8C1C61106FB9623521394" +NETWORK_RPC = "https://testnet-rpc.bitlayer.org" + +agent = AgentSDK(endpoint_uri=NETWORK_RPC, proxy_address=AGENT_PROXY_ADDRESS) +``` + +To create a new agent, call the createAndRegisterAgent method: + +```python +import time +from ai_agent.entities import ( + AgentSettings, + AgentHeader, + MessageType, + Priority +) +from ai_agent.utils import ( + generate_uuid_v4 +) + +AGENT_SETTINGS = AgentSettings( + signers=[ + "0x4b1056f504f32c678227b5Ae812936249c40AfBF", + "0xB973476e0cF88a3693014b99f230CEB5A01ac686", + "0x6cF0803D049a4e8DC01da726A5a212BCB9FAC1a1", + "0x9D46daa26342e9E9e586A6AdCEDaD667f985567B", + "0x33AF673aBcE193E20Ee94D6fBEb30fEf0cA7015b", + "0x868D2dE4a0378450BC62A7596463b30Dc4e3897E", + "0xD4E157c36E7299bB40800e4aE7909DDcA8097f67", + "0xA3866A07ABEf3fD0643BD7e1c32600520F465ca8", + "0x62f642Ae0Ed7F12Bc40F2a9Bf82ccD0a3F3b7531" + ], + threshold=2, + converter_address="0xaB303EF87774D9D259d1098E9aA4dD6c07F69240", + agent_header=AgentHeader( + version="1.0", + message_id="d4d0813f-ceb7-4ce1-8988-12899b26c4b6", + source_agent_id="da70f6b3-e580-470f-b88b-caa5369e7778", + source_agent_name="APRO Pull Mode Agent", + target_agent_id="", + timestamp=int(time.time()), + message_type=MessageType.Event, + priority=Priority.Low, + ttl=60 * 60 + ) +) + +dynamic_setting = AGENT_SETTINGS +dynamic_setting.agent_header.source_agent_id = generate_uuid_v4() +dynamic_setting.agent_header.target_agent_id = generate_uuid_v4() +dynamic_setting.agent_header.message_id = generate_uuid_v4() +user_owner = agent.add_account("0x_user_private_key") +result = agent.create_and_register_agent( + transmitter="", + nonce=None, + settings=AGENT_SETTINGS) +print("created agent:", result) + +``` +To verify the data integrity, call the verify method: + +```python +from ai_agent.entities import ( + AgentMessagePayload, + Proofs, + AgentMetadata, +) + +AGENT_CONTRACT = "0xA1903361Ee8Ec35acC7c8951b4008dbE8D12C155" +AGENT_SETTING_DIGEST = "0x010038164dba6abffb84eb5cb538850d9bc5d8f815149a371069b3255fd177a4" +AGENT_PAYLOAD = AgentMessagePayload( + data="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", + data_hash="0x53d9f133f1265bd4391fcdf89b63424cbcfd316c8448f76cc515647267ac0a8e", + proofs=Proofs( + zk_proof="0x", + merkle_proof="0x", + signature_proof="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", + ), + meta_data=AgentMetadata( + content_type="0x", + encoding="0x", + compression="0x" + ) +) +user_owner = agent.add_account("0x_user_private_key") +result = agent.verify( + transmitter=user_owner, + nonce=None, + agent_contract=AGENT_CONTRACT, + settings_digest=AGENT_SETTING_DIGEST, + payload=AGENT_PAYLOAD +) +print("verify:", result) +``` +For more examples, see the [test cases](https://github.com/APRO-com/ai-agent-sdk-python/tree/main/tests). + + +# Other SDKs + +JAVA: https://github.com/APRO-com/ai-agent-sdk-java + +RUST: https://github.com/APRO-com/ai-agent-sdk-rust + +GOLANG: https://github.com/APRO-com/ai-agent-sdk-go + +# Support + +For issues or feature requests: +1. Check existing documentation +2. Submit a GitHub issue with relevant details +3. Include transaction logs and system info if applicable + +# Contributing + +We welcome pull requests! Refer to the project’s CONTRIBUTING.md and open discussions to coordinate efforts. + +# Credits + +- [APRO](https://www.apro.com/) - Plugin sponsor and partner +- [ai-agent-sdk-js](https://github.com/APRO-com/ai-agent-sdk-js) - Underlying agent SDK +- [ethers.js](https://docs.ethers.io/) - Transaction and contract interaction +- Community contributors for feedback and testing + +For more information about Apro plugin capabilities: + +- [Apro Documentation](https://docs.apro.com/en) + +# License + +This plugin is part of the Eliza project. Refer to the main project repository for licensing details. \ No newline at end of file diff --git a/docs/Siyuan/README.md b/docs/SiYuan/README.md similarity index 93% rename from docs/Siyuan/README.md rename to docs/SiYuan/README.md index fb13da8..ec5cdc6 100644 --- a/docs/Siyuan/README.md +++ b/docs/SiYuan/README.md @@ -2,7 +2,7 @@ ‍ -​![image](assets/image-20250122162731-7wkftbw.png)​ +​![image](https://b3log.org/images/brand/siyuan-128.png)​ --- diff --git a/docs/Siyuan/README_cn.md b/docs/SiYuan/README_cn.md similarity index 92% rename from docs/Siyuan/README_cn.md rename to docs/SiYuan/README_cn.md index b3c7247..c1123bc 100644 --- a/docs/Siyuan/README_cn.md +++ b/docs/SiYuan/README_cn.md @@ -1,6 +1,6 @@ # README_cn -​![image](assets/image-20250122162731-7wkftbw.png)​ +​![image](https://b3log.org/images/brand/siyuan-128.png)​ --- diff --git a/docs/Siyuan/assets/image-20250122162241-32a4oma.png b/docs/SiYuan/assets/image-20250122162241-32a4oma.png similarity index 100% rename from docs/Siyuan/assets/image-20250122162241-32a4oma.png rename to docs/SiYuan/assets/image-20250122162241-32a4oma.png diff --git a/docs/Siyuan/assets/image-20250122162425-wlsgw0u.png b/docs/SiYuan/assets/image-20250122162425-wlsgw0u.png similarity index 100% rename from docs/Siyuan/assets/image-20250122162425-wlsgw0u.png rename to docs/SiYuan/assets/image-20250122162425-wlsgw0u.png diff --git a/docs/Siyuan/assets/image-20250122163007-hkuruoe.png b/docs/SiYuan/assets/image-20250122163007-hkuruoe.png similarity index 100% rename from docs/Siyuan/assets/image-20250122163007-hkuruoe.png rename to docs/SiYuan/assets/image-20250122163007-hkuruoe.png diff --git a/docs/Siyuan/assets/image-20250122162731-7wkftbw.png b/docs/Siyuan/assets/image-20250122162731-7wkftbw.png deleted file mode 100644 index 2bb3189..0000000 Binary files a/docs/Siyuan/assets/image-20250122162731-7wkftbw.png and /dev/null differ diff --git a/docs/Typral/README.md b/docs/Typral/README.md new file mode 100644 index 0000000..30c1f83 --- /dev/null +++ b/docs/Typral/README.md @@ -0,0 +1,11 @@ + + +# [Typral](https://www.typral.com) + +Fast AI writer assistant - Let AI help you quickly improve article, paper, text... + +## UI + + +## 配置 Deepseek API + \ No newline at end of file diff --git a/docs/Typral/README_cn.md b/docs/Typral/README_cn.md new file mode 100644 index 0000000..88d3162 --- /dev/null +++ b/docs/Typral/README_cn.md @@ -0,0 +1,11 @@ + + +# [Typral](https://www.typral.com) + +超快的AI写作助手 - 让AI帮你快速优化日报,文章,文本等等... + +## UI + + +## 配置 Deepseek API + \ No newline at end of file diff --git a/docs/Typral/assets/screenshot1.png b/docs/Typral/assets/screenshot1.png new file mode 100644 index 0000000..a2ad8d4 Binary files /dev/null and b/docs/Typral/assets/screenshot1.png differ diff --git a/docs/Typral/assets/screenshot2.png b/docs/Typral/assets/screenshot2.png new file mode 100644 index 0000000..763c981 Binary files /dev/null and b/docs/Typral/assets/screenshot2.png differ diff --git a/docs/codegate/README.md b/docs/codegate/README.md new file mode 100644 index 0000000..9653112 --- /dev/null +++ b/docs/codegate/README.md @@ -0,0 +1,158 @@ +# CodeGate: secure AI code generation + +CodeGate is a **local gateway** that makes AI agents and coding assistants safer. It +ensures AI-generated recommendations adhere to best practices while safeguarding +your code's integrity and protecting your privacy. With CodeGate, you can +confidently leverage AI in your development workflow without sacrificing +security or productivity. + + + + CodeGate dashboard + + +--- +## ✨ Why choose CodeGate? + +AI coding assistants are powerful, but they can inadvertently introduce risks. +CodeGate protects your development process by: + +- 🔒 Preventing accidental exposure of secrets and sensitive data +- 🛡️ Ensuring AI suggestions follow secure coding practices +- ⚠️ Blocking recommendations of known malicious or deprecated libraries +- 🔍 Providing real-time security analysis of AI suggestions + +--- +## 🚀 Quickstart with 🐋 Deepseek! + +### Prerequisites + +CodeGate is distributed as a Docker container. You need a container runtime like +Docker Desktop or Docker Engine. Podman and Podman Desktop are also supported. +CodeGate works on Windows, macOS, and Linux operating systems with x86_64 and +arm64 (ARM and Apple Silicon) CPU architectures. + +These instructions assume the `docker` CLI is available. If you use Podman, +replace `docker` with `podman` in all commands. + +### Installation + +To start CodeGate, run this simple command (making sure to pass in the +deepseek.com URL as the `CODEGATE_PROVIDER_OPENAI_URL` environment variable): + +```bash +docker run --name codegate -d -p 8989:8989 -p 9090:9090 -p 8990:8990 \ + -e CODEGATE_PROVIDER_OPENAI_URL=https://api.deepseek.com \ + --mount type=volume,src=codegate_volume,dst=/app/codegate_volume \ + --restart unless-stopped ghcr.io/stacklok/codegate:latest +``` + +That’s it! CodeGate is now running locally. + +### Using CodeGate and 🐋 Deepseek within Continue + +To use Continue with CodeGate, open the Continue settings and add +the following configuration: + +```json +{ + "title": "Deepseek-r1", + "provider": "openai", + "model": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B", + "apiKey": "YOUR_DEEPSEEK_API_KEY", + "apiBase": "http://localhost:8989/openai", +} +``` + +Just use Continue as normal, and you know longer have to worry about security +or privacy concerns! + +![continue-image](https://github.com/deepseek/awesome-deepseek-integration/blob/codegate/docs/codegate/assets/continue-screen.png) + + +### Using CodeGate and 🐋 Deepseek with Cline + +To use Cline with CodeGate, open the Cline settings and add +the following configuration: + +![cline-settings](https://github.com/deepseek/awesome-deepseek-integration/blob/codegate/docs/codegate/assets/cline-settings.png) + +Just use Cline as normal, and you know longer have to worry about security +or privacy concerns! + +![cline-image](https://github.com/deepseek/awesome-deepseek-integration/blob/codegate/docs/codegate/assets/cline-screen.png) + +--- +## 🖥️ Dashboard + +CodeGate includes a web dashboard that provides: + +- A view of **security risks** detected by CodeGate +- A **history of interactions** between your AI coding assistant and your LLM + + + + CodeGate dashboard + + +### Accessing the dashboard + +Open [http://localhost:9090](http://localhost:9090) in your web browser to +access the dashboard. + +To learn more, visit the +[CodeGate Dashboard documentation](https://docs.codegate.ai/how-to/dashboard). + +--- +## 🔐 Features + +### Secrets encryption + +CodeGate helps you protect sensitive information from being accidentally exposed +to AI models and third-party AI provider systems by redacting detected secrets +from your prompts using encryption. +[Learn more](https://docs.codegate.ai/features/secrets-encryption) + +### Dependency risk awareness + +LLMs’ knowledge cutoff date is often months or even years in the past. They +might suggest outdated, vulnerable, or non-existent packages (hallucinations), +exposing you and your users to security risks. + +CodeGate scans direct, transitive, and development dependencies in your package +definition files, installation scripts, and source code imports that you supply +as context to an LLM. +[Learn more](https://docs.codegate.ai/features/dependency-risk) + +### Security reviews + +CodeGate performs security-centric code reviews, identifying insecure patterns +or potential vulnerabilities to help you adopt more secure coding practices. +[Learn more](https://docs.codegate.ai/features/security-reviews) + +--- +## 🛡️ Privacy first + +Unlike other tools, with CodeGate **your code never leaves your machine**. +CodeGate is built with privacy at its core: + +- 🏠 **Everything stays local** +- 🚫 **No external data collection** +- 🔐 **No calling home or telemetry** +- 💪 **Complete control over your data** + +--- +## 🛠️ Development + +Are you a developer looking to contribute? Dive into our technical resources: + +- [Development guide](https://github.com/stacklok/codegate/blob/main/docs/development.md) +- [CLI commands and flags](https://github.com/stacklok/codegate/blob/main/docs/cli.md) +- [Configuration system](https://github.com/stacklok/codegate/blob/main/docs/configuration.md) +- [Logging system](https://github.com/stacklok/codegate/blob/main/docs/logging.md) + +--- +## 📜 License + +CodeGate is licensed under the terms specified in the +[LICENSE file](https://github.com/stacklok/codegate/blob/main/LICENSE). diff --git a/docs/codegate/README_cn.md b/docs/codegate/README_cn.md new file mode 100644 index 0000000..10de199 --- /dev/null +++ b/docs/codegate/README_cn.md @@ -0,0 +1,132 @@ +# CodeGate:安全的 AI 代码生成 + +CodeGate 是一个**本地代理**,可以让 AI 代理和编码助手更加安全。它确保 AI 生成的建议遵循最佳实践,同时保护您的代码完整性和隐私。使用 CodeGate,您可以在开发工作流程中自信地利用 AI,而不会牺牲安全性或生产力。 + + + + CodeGate dashboard + + +--- +## ✨ 为什么选择 CodeGate? + +AI 编码助手功能强大,但可能会无意中带来风险。CodeGate 通过以下方式保护您的开发过程: + +- 🔒 防止意外泄露机密和敏感数据 +- 🛡️ 确保 AI 建议遵循安全编码实践 +- ⚠️ 阻止推荐已知的恶意或已弃用的库 +- 🔍 提供 AI 建议的实时安全分析 + +--- +## 🚀 使用 🐋 Deepseek 快速开始! + +### 前提条件 + +CodeGate 以 Docker 容器的形式分发。您需要一个容器运行时,如 Docker Desktop 或 Docker Engine。同时也支持 Podman 和 Podman Desktop。CodeGate 可在 Windows、macOS 和 Linux 操作系统上运行,支持 x86_64 和 arm64(ARM 和 Apple Silicon)CPU 架构。 + +以下说明基于 `docker` CLI 可用的前提。如果您使用 Podman,请在所有命令中将 `docker` 替换为 `podman`。 + +### 安装 + +要启动 CodeGate,运行这个简单的命令(确保将 deepseek.com URL 作为 `CODEGATE_PROVIDER_OPENAI_URL` 环境变量传入): + +```bash +docker run --name codegate -d -p 8989:8989 -p 9090:9090 -p 8990:8990 \ + -e CODEGATE_PROVIDER_OPENAI_URL=https://api.deepseek.com \ + --mount type=volume,src=codegate_volume,dst=/app/codegate_volume \ + --restart unless-stopped ghcr.io/stacklok/codegate:latest +``` + +就是这样!CodeGate 现在在本地运行了。 + +### 在 Continue 中使用 CodeGate 和 🐋 Deepseek + +要在 Continue 中使用 CodeGate,打开 Continue 设置并添加以下配置: + +```json +{ + "title": "Deepseek-r1", + "provider": "openai", + "model": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B", + "apiKey": "YOUR_DEEPSEEK_API_KEY", + "apiBase": "http://localhost:8989/openai", +} +``` + +像往常一样使用 Continue,您不再需要担心安全或隐私问题! + +![continue-image](](https://github.com/deepseek/awesome-deepseek-integration/blob/codegate/docs/codegate/assets/continue-screen.png)) + +### 在 Cline 中使用 CodeGate 和 🐋 Deepseek + +要在 Cline 中使用 CodeGate,打开 Cline 设置并添加以下配置: + +![cline-settings](https://github.com/deepseek/awesome-deepseek-integration/blob/codegate/docs/codegate/assets/cline-settings.png) + +像往常一样使用 Cline,您不再需要担心安全或隐私问题! + +![cline-image](https://github.com/deepseek/awesome-deepseek-integration/blob/codegate/docs/codegate/assets/cline-screen.png) + +--- +## 🖥️ 仪表板 + +CodeGate 包含一个 Web 仪表板,提供: + +- CodeGate 检测到的**安全风险**视图 +- AI 编码助手与 LLM 之间的**交互历史** + + + + CodeGate dashboard + + +### 访问仪表板 + +在您的网络浏览器中打开 [http://localhost:9090](http://localhost:9090) 以访问仪表板。 + +要了解更多信息,请访问 [CodeGate 仪表板文档](https://docs.codegate.ai/how-to/dashboard)。 + +--- +## 🔐 功能 + +### 机密加密 + +CodeGate 通过使用加密对检测到的机密进行编辑,帮助您防止敏感信息意外暴露给 AI 模型和第三方 AI 提供商系统。 +[了解更多](https://docs.codegate.ai/features/secrets-encryption) + +### 依赖风险意识 + +LLM 的知识截止日期通常是几个月甚至几年前。它们可能会建议过时的、易受攻击的或不存在的包(幻觉),使您和您的用户面临安全风险。 + +CodeGate 扫描您作为上下文提供给 LLM 的包定义文件、安装脚本和源代码导入中的直接依赖、传递依赖和开发依赖。 +[了解更多](https://docs.codegate.ai/features/dependency-risk) + +### 安全审查 + +CodeGate 执行以安全为中心的代码审查,识别不安全的模式或潜在的漏洞,帮助您采用更安全的编码实践。 +[了解更多](https://docs.codegate.ai/features/security-reviews) + +--- +## 🛡️ 隐私优先 + +与其他工具不同,使用 CodeGate **您的代码永远不会离开您的机器**。CodeGate 以隐私为核心构建: + +- 🏠 **所有数据均本地存储** +- 🚫 **没有外部数据收集** +- 🔐 **没有回传或遥测** +- 💪 **完全控制您的数据** + +--- +## 🛠️ 开发 + +您是想要贡献的开发者吗?深入了解我们的技术资源: + +- [开发指南](https://github.com/stacklok/codegate/blob/main/docs/development.md) +- [CLI 命令和标志](https://github.com/stacklok/codegate/blob/main/docs/cli.md) +- [配置系统](https://github.com/stacklok/codegate/blob/main/docs/configuration.md) +- [日志系统](https://github.com/stacklok/codegate/blob/main/docs/logging.md) + +--- +## 📜 许可证 + +CodeGate 根据 [LICENSE 文件](https://github.com/stacklok/codegate/blob/main/LICENSE) 中指定的条款获得许可。 \ No newline at end of file diff --git a/docs/codegate/assets/cline-screen.png b/docs/codegate/assets/cline-screen.png new file mode 100644 index 0000000..f59dd29 Binary files /dev/null and b/docs/codegate/assets/cline-screen.png differ diff --git a/docs/codegate/assets/cline-settings.png b/docs/codegate/assets/cline-settings.png new file mode 100644 index 0000000..3b60c5e Binary files /dev/null and b/docs/codegate/assets/cline-settings.png differ diff --git a/docs/codegate/assets/codegate.png b/docs/codegate/assets/codegate.png new file mode 100644 index 0000000..d625d61 Binary files /dev/null and b/docs/codegate/assets/codegate.png differ diff --git a/docs/codegate/assets/continue-screen.png b/docs/codegate/assets/continue-screen.png new file mode 100644 index 0000000..1b93a5d Binary files /dev/null and b/docs/codegate/assets/continue-screen.png differ diff --git a/docs/minuet-ai.nvim/README.md b/docs/minuet-ai.nvim/README.md new file mode 100644 index 0000000..a0386bf --- /dev/null +++ b/docs/minuet-ai.nvim/README.md @@ -0,0 +1,182 @@ + +# Minuet AI + +Minuet AI: Dance with Intelligence in Your Code 💃. + +`Minuet-ai` brings the grace and harmony of a minuet to your coding process. +Just as dancers move during a minuet. + +# Features + +- AI-powered code completion with dual modes: + - Specialized prompts and various enhancements for chat-based LLMs on code completion tasks. + - Fill-in-the-middle (FIM) completion for compatible models (DeepSeek, + Codestral, Qwen, and others). +- Support for multiple AI providers (OpenAI, Claude, Gemini, Codestral, Ollama, and + OpenAI-compatible services). +- Customizable configuration options. +- Streaming support to enable completion delivery even with slower LLMs. +- Support `nvim-cmp`, `blink-cmp`, `virtual text` frontend. + +# Requirements + +- Neovim 0.10+. +- [plenary.nvim](https://github.com/nvim-lua/plenary.nvim) +- optional: [nvim-cmp](https://github.com/hrsh7th/nvim-cmp) +- optional: [blink.cmp](https://github.com/Saghen/blink.cmp) +- An API key for at least one of the supported AI providers + +# Installation + +**Lazy.nvim**: + +```lua +specs = { + { + 'milanglacier/minuet-ai.nvim', + config = function() + require('minuet').setup { + -- Your configuration options here + } + end, + }, + { 'nvim-lua/plenary.nvim' }, + -- optional, if you are using virtual-text frontend, nvim-cmp is not + -- required. + { 'hrsh7th/nvim-cmp' }, + -- optional, if you are using virtual-text frontend, blink is not required. + { 'Saghen/blink.cmp' }, +} +``` + +**Rocks.nvim**: + +`Minuet` is available on luarocks.org. Simply run `Rocks install +minuet-ai.nvim` to install it like any other luarocks package. + +**Setting up with virtual text**: + +```lua +require('minuet').setup { + virtualtext = { + auto_trigger_ft = {}, + keymap = { + -- accept whole completion + accept = '', + -- accept one line + accept_line = '', + -- accept n lines (prompts for number) + -- e.g. "A-z 2 CR" will accept 2 lines + accept_n_lines = '', + -- Cycle to prev completion item, or manually invoke completion + prev = '', + -- Cycle to next completion item, or manually invoke completion + next = '', + dismiss = '', + }, + }, +} +``` + +**Setting up with nvim-cmp**: + +
+ +```lua +require('cmp').setup { + sources = { + { + -- Include minuet as a source to enable autocompletion + { name = 'minuet' }, + -- and your other sources + } + }, + performance = { + -- It is recommended to increase the timeout duration due to + -- the typically slower response speed of LLMs compared to + -- other completion sources. This is not needed when you only + -- need manual completion. + fetching_timeout = 2000, + }, +} + + +-- If you wish to invoke completion manually, +-- The following configuration binds `A-y` key +-- to invoke the configuration manually. +require('cmp').setup { + mapping = { + [""] = require('minuet').make_cmp_map() + -- and your other keymappings + }, +} +``` + +
+ +**Setting up with blink-cmp**: + +
+ +```lua +require('blink-cmp').setup { + keymap = { + -- Manually invoke minuet completion. + [''] = require('minuet').make_blink_map(), + }, + sources = { + -- Enable minuet for autocomplete + default = { 'lsp', 'path', 'buffer', 'snippets', 'minuet' }, + -- For manual completion only, remove 'minuet' from default + providers = { + minuet = { + name = 'minuet', + module = 'minuet.blink', + score_offset = 8, -- Gives minuet higher priority among suggestions + }, + }, + }, + -- Recommended to avoid unnecessary request + completion = { trigger = { prefetch_on_insert = false } }, +} +``` + +
+ +**LLM Provider Examples**: + +**Deepseek**: + +```lua +-- you can use deepseek with both openai_fim_compatible or openai_compatible provider +require('minuet').setup { + provider = 'openai_fim_compatible', + provider_options = { + openai_fim_compatible = { + api_key = 'DEEPSEEK_API_KEY', + name = 'deepseek', + optional = { + max_tokens = 256, + top_p = 0.9, + }, + }, + }, +} + + +-- or +require('minuet').setup { + provider = 'openai_compatible', + provider_options = { + openai_compatible = { + end_point = 'https://api.deepseek.com/v1/chat/completions', + api_key = 'DEEPSEEK_API_KEY', + name = 'deepseek', + optional = { + max_tokens = 256, + top_p = 0.9, + }, + }, + }, +} +``` diff --git a/docs/minuet-ai.nvim/README_cn.md b/docs/minuet-ai.nvim/README_cn.md new file mode 100644 index 0000000..31610dd --- /dev/null +++ b/docs/minuet-ai.nvim/README_cn.md @@ -0,0 +1,172 @@ +# Minuet AI + +Minuet AI：在您的代码中翩翩起舞，挥洒智能 💃。 + +`Minuet-ai` 将小步舞曲的优雅与和谐带入您的编码流程。正如舞者在小步舞曲中舞动一样。 + +# 特性 + +- 基于 AI 的代码补全，提供双重模式： + - 针对代码补全任务，为基于聊天的 LLMs 提供专门的提示和各种增强功能。 + - 针对兼容的模型（DeepSeek、Codestral、Qwen 等）提供中间填充 (FIM) 补全。 +- 支持多种 AI 提供商（OpenAI、Claude、Gemini、Codestral、Ollama 和兼容 OpenAI 的服务）。 +- 可自定义配置选项。 +- 支持流式传输，即使使用较慢的 LLMs 也能实现补全的交付。 +- 支持 `nvim-cmp`、`blink-cmp`、`virtual text` 前端。 + +# 要求 + +- Neovim 0.10+。 +- [plenary.nvim](https://github.com/nvim-lua/plenary.nvim) +- 可选： [nvim-cmp](https://github.com/hrsh7th/nvim-cmp) +- 可选： [blink.cmp](https://github.com/Saghen/blink.cmp) +- 至少一个受支持的 AI 提供商的 API 密钥 + +# 安装 + +**Lazy.nvim：** + +```lua +specs = { + { + 'milanglacier/minuet-ai.nvim', + config = function() + require('minuet').setup { + -- 在此处配置您的选项 + } + end, + }, + { 'nvim-lua/plenary.nvim' }, + -- 可选，如果您使用 virtual-text 前端，则不需要 nvim-cmp。 + { 'hrsh7th/nvim-cmp' }, + -- 可选，如果您使用 virtual-text 前端，则不需要 blink。 + { 'Saghen/blink.cmp' }, +} +``` + +**Rocks.nvim：** + +`Minuet` 可在 luarocks.org 上获取。只需运行 `Rocks install minuet-ai.nvim` 即可像安装其他 luarocks 包一样安装它。 + +**使用 virtual text 进行设置：** + +```lua +require('minuet').setup { + virtualtext = { + auto_trigger_ft = {}, + keymap = { + -- 接受完整补全 + accept = '', + -- 接受一行 + accept_line = '', + -- 接受 n 行（提示输入数字） + -- 例如，“A-z 2 CR”将接受 2 行 + accept_n_lines = '', + -- 切换到上一个补全项，或手动调用补全 + prev = '', + -- 切换到下一个补全项，或手动调用补全 + next = '', + dismiss = '', + }, + }, +} +``` + +**使用 nvim-cmp 进行设置：** + +
+ +```lua +require('cmp').setup { + sources = { + { + -- 包含 minuet 作为源以启用自动补全 + { name = 'minuet' }, + -- 和您的其他来源 + } + }, + performance = { + -- 建议增加超时时间，因为与其他补全来源相比，LLMs 的响应速度通常较慢。如果您只需要手动补全，则不需要此设置。 + fetching_timeout = 2000, + }, +} + + +-- 如果你希望手动调用补全， +-- 以下配置将 `A-y` 键绑定到手动调用配置。 +require('cmp').setup { + mapping = { + [""] = require('minuet').make_cmp_map() + -- 和您的其他键映射 + }, +} +``` + +
+ +**使用 blink-cmp 进行设置：** + +
+ +```lua +require('blink-cmp').setup { + keymap = { + -- 手动调用 minuet 补全。 + [''] = require('minuet').make_blink_map(), + }, + sources = { + -- 启用 minuet 进行自动补全 + default = { 'lsp', 'path', 'buffer', 'snippets', 'minuet' }, + -- 仅对于手动补全，从默认值中删除 'minuet' + providers = { + minuet = { + name = 'minuet', + module = 'minuet.blink', + score_offset = 8, -- 在建议中赋予 minuet 更高的优先级 + }, + }, + }, + -- 建议避免不必要的请求 + completion = { trigger = { prefetch_on_insert = false } }, +} +``` + +
+ +**LLM 提供商示例：** + +**Deepseek：** + +```lua +-- 你可以使用 openai_fim_compatible 或 openai_compatible 提供商来使用 deepseek +require('minuet').setup { + provider = 'openai_fim_compatible', + provider_options = { + openai_fim_compatible = { + api_key = 'DEEPSEEK_API_KEY', + name = 'deepseek', + optional = { + max_tokens = 256, + top_p = 0.9, + }, + }, + }, +} + + +-- 或者 +require('minuet').setup { + provider = 'openai_compatible', + provider_options = { + openai_compatible = { + end_point = 'https://api.deepseek.com/v1/chat/completions', + api_key = 'DEEPSEEK_API_KEY', + name = 'deepseek', + optional = { + max_tokens = 256, + top_p = 0.9, + }, + }, + }, +} +``` diff --git a/docs/tomemo/README.md b/docs/tomemo/README.md index 86c637f..f550279 100644 --- a/docs/tomemo/README.md +++ b/docs/tomemo/README.md @@ -12,7 +12,19 @@ ToMemo is a phrasebook + clipboard history + keyboard iOS app with integrated AI ## Integrate with Deepseek API -Go to Settings-Extensions-AI Services-AI Providers to add the Deepseek API Key. -After adding, you can turn on the 「show in bottom tab」 in the AI service page, so that you can talk to Deepseek directly in the application. +- Go to "Settings-Extensions-AI Services-AI Providers", click "Add" in the top right corner, and select "DeepSeek" in the **Provider** field. +- Enter your API Key in the **API Key** field. +- Click the "Test" button to verify if the input is valid. +- Click "Load Models" to select the model you want to use +- Turn on "Enable" and click "Save" -![image](assets/Integrate.jpg) +![image](assets/app-provider.png) + +## Use + +- Go to "Settings-Extensions-AI Services" +- Click "AI Assistant" to enter the AI Assistant page +- Add an AI Assistant in the top right corner, you can select "Deepseek" in the models +- Start chatting with Deepseek + +![image](assets/use-deepseek.png) diff --git a/docs/tomemo/README_cn.md b/docs/tomemo/README_cn.md index b46b0b2..766675a 100644 --- a/docs/tomemo/README_cn.md +++ b/docs/tomemo/README_cn.md @@ -12,7 +12,19 @@ ToMemo 是一款短语合集 + 剪切板历史 + 键盘输出的 iOS 应用， ## Integrate with Deepseek API -进入设置-扩展-AI 服务-AI 供应商，即可添加 Deepseek API Key。 -添加完成后，可以 AI 服务页面中开启底部 Tab 页，方便应用中直接与 Deepseek 对话。 +- 进入「设置-扩展-AI 服务-AI 供应商」，点击右上角「添加」，在**供应商**中选择「DeepSeek」。 +- 在**API Key**中输入你的 API Key。 +- 点击「测试」按钮，测试填入是否可用。 +- 点击「加载模型」，选择需要使用的模型 +- 打开「启用」后，点击「保存」 -![image](assets/Integrate.jpg) +![image](assets/app-provider.png) + +## Use + +- 进入「设置-扩展-AI 服务」， +- 点击「AI 助手」进入 AI 助手页面， +- 右上角添加 AI 助手，可以在模型中选择「深度求索」 +- 开始和 Deepseek 聊天 + +![image](assets/use-deepseek.png) diff --git a/docs/tomemo/assets/app-provider.png b/docs/tomemo/assets/app-provider.png new file mode 100644 index 0000000..5309a86 Binary files /dev/null and b/docs/tomemo/assets/app-provider.png differ diff --git a/docs/tomemo/assets/use-deepseek.png b/docs/tomemo/assets/use-deepseek.png new file mode 100644 index 0000000..2c0a1a1 Binary files /dev/null and b/docs/tomemo/assets/use-deepseek.png differ
🐠 Abso OpenAIフォーマットを使用するあらゆるLLMプロバイダと対話するためのTypeScript SDK.
Icon siri_deepseek_shortcut