Slothlist: a gentle task planner powered by an open-weight model

Description and use

Slothlist is a low‑effort task planner that turns a messy task dump into a gentle first step, using an open‑weight LLM via Backboard API, while keeping credentials local.

Problem

When a to‑do list feels heavy, deciding what to do next can be overwhelming.

Outcome

Slothlist demonstrates how an open‑weight model can power a low‑effort task planner while keeping credentials local; it requires an internet connection and a Backboard API key.

Workflow

• 1. User enters task text, energy level, and time budget into the browser UI.
• 2. Browser sends this data to a local Python server.
• 3. Server builds a payload and calls Backboard’s Threads Messages API via OpenRouter (default Qwen2.5 7B Instruct).
• 4. Backboard returns a structured JSON plan.
• 5. Browser displays the plan to the user.
• 6. Optional memory preference can be toggled to store planning preferences in Backboard threads.

Evidence

Slothlist turns a messy task dump into one gentle first step, a few small moves, and permission to let the rest wait.

Source

The browser sends the task text, energy level, time budget, and optional memory preference to a Python server running on your own machine.

Source

The server calls Backboard's Threads Messages API and asks for a structured JSON plan. The default route is OpenRouter with Qwen2.5 7B Instruct.

Source

The Backboard API key stays in a local, Git-ignored .env file and is read by the server. It is never placed in the browser JavaScript.

Source

The default model route is OpenRouter's Qwen2.5 7B Instruct, an open-weight model.

Source