Aalsi: let a local model dig the deadlines out of your group chats

Description and use

Aalsi is a local-first deadline catcher for college WhatsApp groups that extracts due dates from exported chats, provides a daily digest, missed‑deadline list, and .ics calendar, all running locally without external services.

Problem

College WhatsApp groups bury important deadline messages under many unrelated messages, making it hard for students to find due dates.

Outcome

Aalsi provides a concise daily summary of upcoming deadlines, a missed‑deadline list, and a calendar file, all derived from a local model run on the user’s machine, preserving privacy and eliminating API costs.

Workflow

• Export WhatsApp group chat as .txt (without media).
• Drop the .txt file into Aalsi’s interface.
• Aalsi parses the chat using a Go parser and extracts deadlines via a local Gemma model or rule‑based fallback.
• It resolves dates deterministically in Go, tracks updates and postponements, and generates a morning digest, missed‑deadline list, and .ics calendar file.
• Items with low confidence are sent to a review queue where the user can confirm or dismiss them.
• All data remains on the local machine and can be purged via a button.

Evidence

Every college WhatsApp group has the same problem: one message that matters buried under three hundred memes and "ok sir"s.

Source

Aalsi runs on your own machine by design, so there is no hosted demo.

Source

Aalsi extracts deadlines using open-weight Gemma running locally on Ollama, resolves dates deterministically in Go, tracks living deadline state across reminders and updates, and delivers a morning digest, a "you may have missed" list, a calendar file, and a review queue.

Source

The only service it talks to is Ollama on localhost.

Source

I took them in the rule-based mode, so each one shows "Extracted by: rule mode".

Source