Getting Started with SafeAgent

SafeAgent is a self-hosted AI agent framework that runs entirely on your machine. It provides a chat interface where you can give natural language commands, and your AI agent executes real actions — reading email, scraping websites, posting to Slack, managing GitHub repos, and much more.

What You Can Do

• Chat with AI — natural language interface, like ChatGPT but connected to your real services
• Read & send email — works with Gmail, Outlook, Yahoo, or any IMAP/SMTP server with AI-powered triage
• Knowledge Base (RAG) — upload PDFs, docs, CSVs and ask AI questions grounded in your data
• MCP Protocol — connect to any MCP server (Google Drive, Salesforce, databases) without custom code
• Multi-step Workflows — chain actions together: read email → AI triage → post to Slack
• Event Watchers — react to new emails, GitHub issues, and PRs automatically
• Inbound Webhooks — receive events from GitHub, Slack, Stripe and process them with AI
• Agent Memory — your AI remembers preferences and context across sessions
• Multi-Agent — specialist agents that delegate, pipeline, and triage together
• Voice Input — talk to your agent via voice, with Whisper transcription
• Scrape any website — extract text, links, prices, emails from any URL
• Manage GitHub — list repos, create issues, track PRs
• Post to Slack — read channels, send messages
• Control via Telegram/Discord — send tasks from your phone
• API Key access — programmatic access to all 113 endpoints

Key Principles

• Self-hosted — everything runs on your machine, nothing goes to the cloud
• Zero config to start — ships with local AI (Ollama + SafeAgent 7B), no API keys needed
• Encrypted — all credentials encrypted with AES-256-GCM at rest
• Judgment engine — every action is risk-scored; dangerous actions require approval
• UI-configured — all services connected through the dashboard, no .env editing for integrations

Requirements