Overview
Helmis is a self-hosted autonomous AI executive secretary for WhatsApp, powered by Google Gemini, multi-step ReAct tool calling, localized semantic memory, and a document vault. It is described as a private, zero-latency AI Executive Secretary built for real-world personal coordination over WhatsApp. It supports Personal Solo Mode (1 user) and Duo Mode for couples or partners in a shared group chat.
The system manages schedules, tasks, contacts, shared notes, an encrypted Document Vault, PDF manipulation and text extraction, live web search, dynamic media dispatching, and proactive deadline reminders. It includes forced tool calling and an anti-promissory guardrail, procedural memory and self-learning skills, autonomous auto-crystallization, a universal code execution sandbox, an autonomous ReAct loop and cascade, mid-turn mailbox steering, multimodal intelligence and vision OCR, a categorized document vault, a Google Workspace reader, proactive cron and exact-second timers, a single source of truth system prompt, and zero AI slop.
Setup involves cloning the repository, configuring environment variables, starting services with Docker Compose, and connecting WhatsApp through the WAHA dashboard. A test suite is available with 21 modules and 179 test cases. The project is licensed under the MIT License.
Pricing
Free
Detailed plans have not been confirmed in our catalog. Check the official website for current limits and billing terms.
Visit WebsitePrices and limits may change. Confirm the currency, billing period, seat minimum and usage caps on the official website.
Use Cases
- Users use Helmis to manage schedules, tasks, contacts, shared notes, and proactive deadline reminders over WhatsApp.
- Users use Helmis to store and dispatch categorized documents from a local persistent vault.
- Users use Helmis to extract text from scanned PDFs, diagrams, spreadsheets, and presentation slides.
- Users use Helmis to inspect published Google Sheets, Docs, Slides, and Drive documents.
- Users use Helmis to execute Python 3 for calculations, date/time manipulation in WIB, and data processing.
- Users use Helmis to learn and persist new operational playbooks as SKILL.md files.
- Users use Helmis to receive proactive reminders via cron ticks and exact-second timers.
- Users use Helmis to start services with Docker Compose.
Features
AI assistant
Natural language interface
Workflow automation
Reviews
No reviews yet. Be the first to share your experience!