[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-118":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":9,"language":10,"languages":9,"totalLinesOfCode":9,"stars":11,"forks":12,"watchers":13,"openIssues":14,"contributorsCount":15,"subscribersCount":15,"size":15,"stars1d":14,"stars7d":16,"stars30d":17,"stars90d":15,"forks30d":15,"starsTrendScore":18,"compositeScore":19,"rankGlobal":9,"rankLanguage":9,"license":9,"archived":20,"fork":20,"defaultBranch":21,"hasWiki":22,"hasPages":20,"topics":23,"createdAt":9,"pushedAt":9,"updatedAt":24,"readmeContent":25,"aiSummary":26,"trendingCount":15,"starSnapshotCount":15,"syncStatus":27,"lastSyncTime":28,"discoverSource":29},118,"evonic","anvie\u002Fevonic","anvie","Open Agentic AI Platform - The home your agents deserve",null,"Python",261,56,4,3,0,18,47,9,5.27,false,"main",true,[],"2026-06-12 02:00:08","# Evonic\n\n> **Design. Deploy. Orchestrate.**  \n> *Your Models. Your Rules. Your Swarm.*\n\nEvonic is an **agentic AI framework** for designing, building, and orchestrating intelligent agents from concept to production. It empowers you to define every aspect of an agent — its **model**, **tools**, **knowledge base**, **channels**, and **skills** — and compose them into multi-agent systems that operate autonomously across distributed environments.\n\n**Full documentation:** [evonic.dev](https:\u002F\u002Fevonic.dev)\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"static\u002Fimg\u002Fmascot-large.png\" alt=\"Evonic Logo\" width=\"200\">\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"static\u002Fimg\u002Fevonic-web.jpg\" alt=\"Evonic Web UI Screenshot\" width=\"600\">\n\u003C\u002Fp>\n\n---\n\n## Three Core Differentiators\n\nEvonic is not just another agent framework. Three architectural decisions set it apart:\n\n### 1. Workplace — Anywhere Execution\n\nAgents are not tied to a single machine. A **Workplace** is a first-class execution environment that can be:\n\n- **Local** — sandboxed workspace on the host machine\n- **Remote** — SSH servers, edge devices, or any machine with network access\n- **Cloud** — lightweight Evonet connector that requires no public IP, no SSH, and no firewall rules\n\nThis means your agents can operate across your entire infrastructure — development laptops, production servers, and cloud instances — with a single abstraction layer.\n\n### 2. Agent-to-Agent Communication\n\nCommunication between agents is a first-class protocol, not an afterthought. Agents can message, delegate, and coordinate with each other natively. This enables:\n\n- **Multi-agent swarms** where each agent has a distinct role and toolset\n- **Hierarchical orchestration** with supervisor agents managing worker agents\n- **Peer-to-peer collaboration** for complex multi-step workflows\n\nEach agent maintains its own identity, state, and capabilities, making swarm intelligence a natural pattern rather than a bolt-on feature.\n\n### 3. Heuristic Mal-activity Detection System\n\nSafety is not optional. Every action an agent takes is inspected through a real-time, multi-layer heuristic detection system that identifies and blocks dangerous patterns before execution. The system monitors for:\n\n- Mass file deletion and privilege escalation attempts\n- Unauthorized remote code execution\n- Behavioral drift beyond expected action boundaries\n\nWhen suspicious activity is detected, the system escalates to a human operator rather than blindly executing — giving you a safety net that enables genuine agent autonomy without compromising security.\n\n---\n\n## Key Features\n\n| Feature | Description |\n|---------|-------------|\n| **Agents** | Independent, LLM-powered assistants with custom tools, knowledge bases, and isolated workspaces |\n| **Models** | Pluggable LLM backends — any OpenAI-compatible API, local or cloud |\n| **Skills** | Installable packages that bundle tool definitions with Python backends |\n| **Plugins** | Event-driven extensions for custom integrations and background workers |\n| **Workplaces** | Execution environments: local directories, SSH servers, or cloud devices via Evonet |\n| **Evonet** | Lightweight Go connector for remote execution without SSH or firewall rules |\n| **Scheduler** | Cron-based triggers, recurring tasks, and reminders for agents |\n| **Channels** | Connect agents to Telegram, WhatsApp, Discord, Slack, and custom interfaces |\n| **Evaluation Engine** | Automated LLM evaluation with customizable regex and heuristic evaluators |\n\n---\n\n## Getting Started\n\n### Prerequisites\n\n- **Python 3.8+**\n- **LLM endpoint** — any OpenAI-compatible API (local or cloud)\n\n### Installation\n\nOne-liner install:\n\n```bash\ncurl -fsSL https:\u002F\u002Fevonic.dev\u002Finstall.sh | bash\n```\n\nThis clones the repository, sets up a virtual environment, installs dependencies, generates configuration, and guides you through adding `evonic` to your PATH.\n\n**Manual installation:**\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fanvie\u002Fevonic\ncd evonic\npip install -r requirements.txt\nchmod +x .\u002Fevonic\n```\n\n### Start\n\n```bash\n.\u002Fevonic start\n```\n\nOpen `http:\u002F\u002Flocalhost:8080` in your browser.\n\n### Docker Sandbox (optional)\n\nAgent tools like `bash` and `runpy` execute inside an isolated Docker container by default:\n\n```bash\ndocker build -t evonic-sandbox:latest docker\u002Ftools\u002F\n```\n\nConfigure resource limits in `.env` (memory, CPU, network). If Docker is unavailable, set `sandbox_enabled=0` to fall back to local execution.\n\n---\n\n## Agents\n\nEach agent is designed from the ground up with six configurable dimensions:\n\n| Dimension | Description |\n|-----------|-------------|\n| **Concept** | System prompt and identity — who the agent is and how it behaves |\n| **Model** | LLM backend — which model powers the agent's reasoning |\n| **Tools** | Capabilities — what actions the agent can take |\n| **Knowledge Base** | Reference documents — what information the agent can access |\n| **Channels** | Interfaces — where users interact with the agent |\n| **Skills** | Modular extensions — additional capabilities installed on demand |\n\nCreate and manage agents via the web UI (`\u002Fagents`) or CLI:\n\n```bash\n.\u002Fevonic agent add my_bot --name \"My Bot\"\n.\u002Fevonic agent add dev_bot --name \"Dev Bot\" --skillset coder\n.\u002Fevonic agent enable my_bot\n.\u002Fevonic agent remove my_bot\n```\n\n---\n\n## Channels\n\nConnect your agents to the platforms your users already use:\n\n| Channel | Status | Library |\n|---------|--------|---------|\n| **Telegram** | ✅ Implemented | `python-telegram-bot` |\n| **WhatsApp** | ✅ Implemented | `@whiskeysockets\u002Fbaileys` (Node.js sidecar) |\n| **Discord** | 🔄 Planned | `discord.py` |\n| **Slack** | 🔄 Planned | `slack-sdk` |\n\n---\n\n## Skills\n\nSkills extend agents with new capabilities. Install via CLI:\n\n```bash\n.\u002Fevonic skill install path\u002Fto\u002Fskill.zip\n.\u002Fevonic skill list\n.\u002Fevonic skill enable math\n.\u002Fevonic skill uninstall math\n```\n\nSkills follow a **load → context → execute** lifecycle, keeping the agent's system prompt lean and modular.\n\n---\n\n## Plugins\n\nPlugins are event-driven extensions that hook into Evonic's event stream. Manage them via CLI:\n\n```bash\n.\u002Fevonic plugin install path\u002Fto\u002Fplugin.zip\n.\u002Fevonic plugin list\n.\u002Fevonic plugin uninstall my_plugin\n```\n\n---\n\n## Models\n\nManage LLM configurations:\n\n```bash\n.\u002Fevonic model add gpt4o --name \"GPT-4o\" --provider openai --api-key \"sk-...\" --base-url \"https:\u002F\u002Fapi.openai.com\u002Fv1\"\n.\u002Fevonic model list\n.\u002Fevonic model rm gpt4o\n```\n\n---\n\n## Use Cases\n\nEvonic's architecture unlocks a broad spectrum of real-world applications. Here are fifteen concrete scenarios:\n\n### Customer Service\nDeploy agents that handle support tickets, answer FAQs, process refunds, and escalate complex issues — all within your existing Telegram or WhatsApp channels.\n\n### Personal Companion\nBuild personal assistants that manage daily tasks, set reminders, conduct research, and maintain long-term context across conversations.\n\n### Agentic Swarm \u002F Multi-Agent Orchestration\nOrchestrate multiple agents with distinct roles — a researcher, a writer, a reviewer, and a publisher — collaborating autonomously on complex deliverables.\n\n### Automation & DevOps\nDeploy agents that monitor server health, trigger deployments, roll back faulty releases, and respond to incidents with automated runbooks.\n\n### Research Assistant\nCreate agents that perform literature reviews, extract structured data from documents, summarize findings, and generate citations.\n\n### Customer Onboarding\nGuide new users through product setup with interactive agents that adapt to each user's pace and knowledge level.\n\n### Quality Assurance & Evaluation\nAutomate LLM evaluation pipelines — define test cases, run evaluations across models, and generate benchmark reports automatically.\n\n### Internal Helpdesk\nProvide IT support, HR policy lookups, and facility requests through a single agent interface connected to your internal knowledge base.\n\n### E-commerce Assistant\nPower product recommendations, order tracking, cancellation requests, and inventory inquiries — connected to your commerce backend.\n\n### Healthcare Triage\nDeploy agents that conduct initial symptom assessment, schedule appointments, and route critical cases to the appropriate specialist.\n\n### Education Tutor\nBuild adaptive tutoring agents that personalize learning paths, grade assignments, and provide real-time feedback to students.\n\n### Content Moderation\nScan user-generated content for harmful patterns, flag violations, and take appropriate action — all within configurable safety boundaries.\n\n### Financial Advisory\nCreate agents that analyze portfolios, generate market summaries, assess risk profiles, and provide data-driven financial insights.\n\n### Agentic ERP\nOrchestrate enterprise resource planning workflows — supply chain monitoring, inventory optimization, procurement automation, and financial reconciliation — through specialized agents that coordinate across departments.\n\n### AI Workflow Orchestration\nDesign end-to-end AI pipelines where agents manage the entire lifecycle: data ingestion, preprocessing, model training, evaluation, and deployment — with each stage handled by a specialized agent.\n\n---\n\n## Architecture Overview\n\n```\nUser Message\n    ↓\nChannel (Telegram, Web, WhatsApp, etc.)\n    ↓\nAgent Runtime\n    ├── Load agent config (system prompt, model, tools)\n    ├── Load\u002Fcreate session (per-user persistence)\n    ├── Build messages (system prompt + history + new message)\n    ├── Call LLM\n    ├── Execute tool calls (if any)\n    ├── Heuristic safety check on every action\n    └── Loop until final response\n    ↓\nResponse → Channel → User\n```\n\n---\n\n## License\n\nEvonic is open source. See the [LICENSE](LICENSE) file for details.\n\n---\n\n*Built with ❤️ by [Robin Syihab](https:\u002F\u002Fgithub.com\u002Fanvie)*\n","Evonic 是一个开放的智能代理 AI 平台，旨在帮助用户从概念到生产全过程设计、构建和编排智能代理。它支持定义代理的模型、工具、知识库、渠道和技能，并将这些组件组合成能够在分布式环境中自主运行的多代理系统。其核心功能包括：任何地方执行的工作场所，支持本地、远程及云端环境；原生的代理间通信机制，促进多代理协作与分层管理；以及基于启发式的恶意活动检测系统，确保操作安全。适用于需要构建跨平台且高度可扩展的智能代理应用的场景，如企业级自动化流程、复杂任务协调等。",2,"2026-06-11 02:30:58","CREATED_QUERY"]