Sloth Desk: Paste a Message You’re Avoiding, Get a Checklist (Local Gemma, No Cloud)

Description and use

Sloth Desk is a local, privacy‑preserving tool that uses the open‑weight Gemma 3 4B model via Ollama to convert long messages into concise checklists of dates, deadlines, amounts, and requirements, enabling users to quickly see all critical details without uploading data to the cloud.

Problem

People keep putting off reading long, wordy notices such as college placement alerts, landlord messages, or bills, risking missed deadlines and important details.

Outcome

Sloth Desk reduces the time to extract actionable items from long notices from several minutes of reading to 30–60 seconds of processing, capturing roughly 95 % of details in tests.

Workflow

• 1. Install Ollama and pull the Gemma 3 4B model via ollama pull gemma3:4b.
• 2. Download index.html from the repo.
• 3. Serve the file locally with python -m http.server 8000 to allow browser access to Ollama.
• 4. Open http://localhost:8000/index.html, paste a message or drag a .txt file, and click "Decode it".
• 5. The page calls Ollama’s /api/chat with strict JSON output and temperature 0, then renders the checklist.
• 6. The script performs a self‑check: it scans the original text for dates, times, amounts, sizes, and places, verifies each appears in the output, and retries once if any are missing, otherwise flags them in red.

Evidence

Runs fully on your own computer with a local open-weight model (Gemma 3 4B via Ollama). Nothing is uploaded anywhere.

Source

In my testing, Sloth Desk caught roughly 95% of the details in messages.

Source

The page scans the original message for dates, times, amounts, sizes and file formats and checks each one made it into the output.

Source