{"id":45106572,"url":"https://github.com/crystal-autobot/autobot","last_synced_at":"2026-03-04T22:01:43.580Z","repository":{"id":338252852,"uuid":"1156577017","full_name":"crystal-autobot/autobot","owner":"crystal-autobot","description":"Ultra-efficient personal AI assistant powered by Crystal","archived":false,"fork":false,"pushed_at":"2026-03-01T19:00:48.000Z","size":2421,"stargazers_count":28,"open_issues_count":1,"forks_count":2,"subscribers_count":3,"default_branch":"main","last_synced_at":"2026-03-01T20:54:40.675Z","etag":null,"topics":["agent-framework","agentic-ai","crystal-lang"],"latest_commit_sha":null,"homepage":"https://crystal-autobot.github.io/autobot/","language":"Crystal","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/crystal-autobot.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":"docs/security.md","support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-02-12T20:05:57.000Z","updated_at":"2026-03-01T19:00:50.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/crystal-autobot/autobot","commit_stats":null,"previous_names":["crystal-autobot/autobot"],"tags_count":10,"template":false,"template_full_name":null,"purl":"pkg:github/crystal-autobot/autobot","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/crystal-autobot%2Fautobot","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/crystal-autobot%2Fautobot/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/crystal-autobot%2Fautobot/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/crystal-autobot%2Fautobot/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/crystal-autobot","download_url":"https://codeload.github.com/crystal-autobot/autobot/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/crystal-autobot%2Fautobot/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":30095693,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-03-04T21:59:23.547Z","status":"ssl_error","status_checked_at":"2026-03-04T21:57:50.415Z","response_time":59,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["agent-framework","agentic-ai","crystal-lang"],"created_at":"2026-02-19T22:01:26.478Z","updated_at":"2026-03-04T22:01:43.565Z","avatar_url":"https://github.com/crystal-autobot.png","language":"Crystal","funding_links":[],"categories":["Setup Methods (1-10 Minutes)","Main Projects","By Category"],"sub_categories":["Recommended","Other Languages"],"readme":"\u003cp align=\"center\"\u003e\n  \u003cimg src=\"docs/assets/banner-circuit-hex.svg\" alt=\"crystal-autobot\" width=\"100%\"\u003e\n\u003c/p\u003e\n\n\u003cp align=\"center\"\u003e\u003cb\u003eUltra-efficient personal AI assistant powered by Crystal\u003c/b\u003e\u003c/p\u003e\n\n\u003cp align=\"center\"\u003e2MB binary · ~5MB RAM · \u003c20ms startup · Zero runtime dependencies\u003c/p\u003e\n\n## Why Autobot?\n\nInspired by [OpenClaw](https://openclaw.ai/) — rebuilt in [Crystal](https://crystal-lang.org) with security and efficiency first.\n\n2.0MB binary, ~5MB RAM, boots in under 20ms, zero runtime dependencies. Run dozens of bots on a single machine — each with its own personality, workspace, and config.\n\n## ✨ Features\n\n- **🤖 Multi-Provider LLM** — Anthropic, OpenAI, DeepSeek, Groq, Gemini, OpenRouter, AWS Bedrock, vLLM\n- **💬 Chat Channels** — Telegram, Slack, WhatsApp, Zulip with allowlists and custom slash commands\n- **👁️ Vision** — Send photos via Telegram and get AI-powered image analysis\n- **🎤 Voice** — Voice messages auto-transcribed via Whisper (Groq/OpenAI)\n- **🔒 Kernel Sandbox** — Docker/bubblewrap OS-level isolation, not regex path checks\n- **🧠 Memory** — JSONL sessions with consolidation and persistent long-term memory\n- **⏰ Cron** — Cron expressions, intervals, one-time triggers, per-owner isolation\n- **🔧 Extensible** — Plugins, bash auto-discovery, markdown skills, subagents\n- **📊 Observable** — Token tracking, credential sanitization, audit trails\n- **🏃 Multi-Bot** — Isolated directories per bot, run dozens on one machine\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"docs/assets/demo-telegram.jpg\" alt=\"Telegram Chat\" width=\"26%\"\u003e\n  \u003cimg src=\"docs/assets/demo-terminal.png\" alt=\"Autobot Terminal\" width=\"73%\"\u003e\n\u003c/p\u003e\n\n### 🛡️ Production-Grade Security\n\nAutobot uses **kernel-enforced sandboxing** via Docker or bubblewrap — not application-level validation. When the LLM executes commands:\n\n- ✅ **Only workspace directory is accessible** (enforced by Linux mount namespaces)\n- ✅ **Everything else is invisible** to the LLM — your `/home`, `/etc`, system files simply don't exist in the sandbox\n- ✅ **No symlink exploits, TOCTOU, or path traversal** — kernel guarantees workspace isolation\n- ✅ **Process isolation** — LLM can't see or interact with host processes\n- ✅ **Auto-detected** — Uses Docker (macOS/production) or bubblewrap (Linux/dev)\n\n**Example:** When LLM tries `ls ../`, it fails at the OS level because parent directories aren't mounted. No regex patterns, no validation bypasses — just kernel namespaces.\n\n**→ [Security architecture](https://crystal-autobot.github.io/autobot/security/)**\n\n## 🚀 Quick Start\n\n### 1. Install\n\n```bash\n# macOS (Homebrew)\nbrew tap crystal-autobot/tap\nbrew install autobot\n\n# Linux/macOS - Download binary\ncurl -L \"https://github.com/crystal-autobot/autobot/releases/latest/download/autobot-$(uname -s | tr '[:upper:]' '[:lower:]')-$(uname -m)\" -o autobot\nchmod +x autobot\nsudo mv autobot /usr/local/bin/\n\n# Or build from source\ngit clone https://github.com/crystal-autobot/autobot.git\ncd autobot\nmake release\nsudo install -m 0755 bin/autobot /usr/local/bin/autobot\n\n# Or use Docker (multi-arch: amd64, arm64)\ndocker pull ghcr.io/crystal-autobot/autobot:latest\n```\n\n### 2. Create a new bot\n\n```bash\nautobot new optimus\ncd optimus\n```\n\nThis creates an `optimus/` directory with everything you need:\n\n```\noptimus/\n├── .env              # API keys (add yours here)\n├── .gitignore        # Excludes secrets, sessions, logs\n├── config.yml        # Configuration (references .env vars)\n├── sessions/         # Conversation history\n├── logs/             # Application logs\n└── workspace/        # Sandboxed LLM workspace\n    ├── AGENTS.md     # Agent instructions\n    ├── SOUL.md       # Personality definition\n    ├── USER.md       # User preferences\n    ├── memory/       # Long-term memory\n    └── skills/       # Custom skills\n```\n\n### 3. Configure\n\nEdit `.env` and add your API keys:\n\n```bash\nANTHROPIC_API_KEY=sk-ant-...\n```\n\nThe generated `config.yml` references these via `${ENV_VAR}` — no secrets in config files.\n\n### 4. Run\n\n```bash\n# Validate configuration\nautobot doctor\n\n# Interactive mode\nautobot agent\n\n# Single command\nautobot agent -m \"Summarize this project\"\n\n# Gateway (all channels)\nautobot gateway\n```\n\nAutobot automatically detects and logs the sandbox method on startup — Docker on macOS/production, bubblewrap on Linux.\n\n**→ [Full quick start guide](https://crystal-autobot.github.io/autobot/quickstart/)**\n\n## 📚 Documentation\n\n- **Getting started** — [Quick start](https://crystal-autobot.github.io/autobot/quickstart/) · [Configuration](https://crystal-autobot.github.io/autobot/configuration/)\n- **Providers** — [Anthropic](https://crystal-autobot.github.io/autobot/anthropic/) · [OpenAI](https://crystal-autobot.github.io/autobot/openai/) · [DeepSeek](https://crystal-autobot.github.io/autobot/deepseek/) · [Groq](https://crystal-autobot.github.io/autobot/groq/) · [Gemini](https://crystal-autobot.github.io/autobot/gemini/) · [OpenRouter](https://crystal-autobot.github.io/autobot/openrouter/) · [Bedrock](https://crystal-autobot.github.io/autobot/bedrock/) · [vLLM](https://crystal-autobot.github.io/autobot/vllm/)\n- **Channels** — [Telegram](https://crystal-autobot.github.io/autobot/telegram/) · [Slack](https://crystal-autobot.github.io/autobot/slack/) · [Zulip](https://crystal-autobot.github.io/autobot/zulip/)\n- **Features** — [Cron](https://crystal-autobot.github.io/autobot/cron/) · [Media \u0026 voice](https://crystal-autobot.github.io/autobot/media/) · [Web search](https://crystal-autobot.github.io/autobot/web-search/) · [MCP servers](https://crystal-autobot.github.io/autobot/mcp/) · [Memory](https://crystal-autobot.github.io/autobot/memory/) · [Plugins](https://crystal-autobot.github.io/autobot/plugins/)\n- **Security** — [Security guide](https://crystal-autobot.github.io/autobot/security/) · [Sandboxing](https://crystal-autobot.github.io/autobot/sandboxing/)\n- **Operations** — [Deployment](https://crystal-autobot.github.io/autobot/deployment/) · [Architecture](https://crystal-autobot.github.io/autobot/architecture/) · [Development](https://crystal-autobot.github.io/autobot/development/)\n\n**→ [Full documentation](https://crystal-autobot.github.io/autobot/)**\n\n## 💡 Examples\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cb\u003eTelegram Bot with Custom Commands\u003c/b\u003e\u003c/summary\u003e\n\n```yaml\nchannels:\n  telegram:\n    enabled: true\n    token: \"BOT_TOKEN\"\n    allow_from: [\"your_username\"]\n    custom_commands:\n      macros:\n        summarize: \"Summarize our conversation in 3 bullet points\"\n        translate:\n          prompt: \"Translate the following to English\"\n          description: \"Translate text to English\"\n      scripts:\n        deploy:\n          path: \"/home/user/scripts/deploy.sh\"\n          description: \"Deploy to production\"\n        status: \"/home/user/scripts/system_status.sh\"\n```\n\nUse `/summarize` or `/deploy` in Telegram to trigger them.\nCommands with a `description` show it in Telegram's command menu; otherwise the command name is used.\n\n\u003c/details\u003e\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cb\u003eCron Scheduler\u003c/b\u003e\u003c/summary\u003e\n\n```bash\n# Daily morning greeting\nautobot cron add --name \"morning\" \\\n  --message \"Good morning! Here's today's summary\" \\\n  --cron \"0 9 * * *\"\n\n# Hourly reminder\nautobot cron add --name \"reminder\" \\\n  --message \"Stand up and stretch!\" \\\n  --every 3600\n\n# One-time meeting notification\nautobot cron add --name \"meeting\" \\\n  --message \"Team sync in 5 minutes!\" \\\n  --at \"2025-03-01T10:00:00\"\n```\n\n\u003c/details\u003e\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cb\u003eMulti-Provider Setup\u003c/b\u003e\u003c/summary\u003e\n\n```yaml\nproviders:\n  anthropic:\n    api_key: \"${ANTHROPIC_API_KEY}\"\n  openai:\n    api_key: \"${OPENAI_API_KEY}\"\n  deepseek:\n    api_key: \"${DEEPSEEK_API_KEY}\"\n  vllm:\n    api_base: \"http://localhost:8000\"\n    api_key: \"token\"\n\nagents:\n  defaults:\n    model: \"anthropic/claude-sonnet-4-5\"\n    max_tokens: 8192\n    temperature: 0.7\n```\n\n\u003c/details\u003e\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cb\u003eMCP Server Integration\u003c/b\u003e\u003c/summary\u003e\n\nConnect external tools via MCP (Model Context Protocol):\n\n```yaml\nmcp:\n  servers:\n    github:\n      command: \"npx\"\n      args: [\"-y\", \"@modelcontextprotocol/server-github\"]\n      env:\n        GITHUB_TOKEN: \"${GITHUB_TOKEN}\"\n    garmin:\n      command: \"uvx\"\n      args: [\"--python\", \"3.12\", \"--from\", \"git+https://github.com/Taxuspt/garmin_mcp\", \"garmin-mcp\"]\n      env:\n        GARMIN_EMAIL: \"${GARMIN_EMAIL}\"\n```\n\nTools are auto-discovered and available as `mcp_github_*`, `mcp_garmin_*`, etc.\n\n```bash\nautobot agent -m \"list my recent garmin activities\"\nautobot agent -m \"show open issues in crystal-autobot/autobot\"\n```\n\n\u003c/details\u003e\n\n## 🔧 Development\n\n### Prerequisites\n- [Crystal](https://crystal-lang.org/install/) \u003e= 1.10.0\n\n### Commands\n\n```bash\nmake build          # Debug binary\nmake release        # Optimized binary (~2MB)\nmake test           # Run test suite\nmake lint           # Run ameba linter\nmake format         # Format code\n\nmake docker         # Build Docker image\nmake release-all    # Cross-compile for all platforms\nmake help           # Show all targets\n```\n\n**→ [Development guide](https://crystal-autobot.github.io/autobot/development/)**\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcrystal-autobot%2Fautobot","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fcrystal-autobot%2Fautobot","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcrystal-autobot%2Fautobot/lists"}