Turn Telegram messages into Todoist tasks
They use Telegram, an LLM, and n8n to convert plain messages into Todoist todos.
Agent Uses · 1,321 real examples
Household paperwork, appointments, home projects and the small chores that eat a week. Each entry shows what the person asked for, how it went, and a prompt to try it yourself.
The day's AI news plus 3 agent uses you can try, with ready-to-paste prompts. Free; unsubscribe anytime.
They use Telegram, an LLM, and n8n to convert plain messages into Todoist todos.
A developer let OpenClaw turn Telegram messages into additions on a task list that lived on their home computer.
They used an AI agent to change Home Assistant rules so harsh LEDs give way to softer strips, with a late-night shutdown and red sleep mode.
A family organizer can turn forwarded school messages and flyer images into calendar entries and reminders.
A developer kept a Claude Code agent persistent across sessions by loading memory and identity at start and saving updates at the end.
OpenClaw used a user’s past chats, search activity, and Instagram trail to pick restaurants, share daily suggestions, and draft memory edits for review.
A non-coder used Codex to make a local meal-planning app that draws on kitchen stock, garden harvests, dietary limits, appliances, and the forecast.
Use Alexa+ to summarize handwritten Scribe notes and draft to-dos, events, and reminders.
Karpathy set up Dobby to handle home devices through plain language instead of juggling separate apps.
A user let Claude computer use search and submit applications, and it finished 50 job applications in under 30 minutes.
An agent explored a home network, decoded several device interfaces, and assembled one place to run the whole house.
With browser access, an agent created materials, searched job boards, and sent out more than 100 applications in three days.
They built a Raspberry Pi setup that watches sleeping posture and flips a fan on or off with a remote button pusher.
A custom assistant can pull files from your device, fill visa applications, upload documents, and pause for review before the final step.
A developer fed Claude Code genetic, lab, scan, and supplement data, then spent two hours getting a plan tied to those inputs.
Felix Rieseberg had Claude Cowork transfer details from hospital PDFs into his insurance forms while he was busy with newborn care.
A family used a custom agent to capture chores, calls, and schedule changes from spoken requests and keep everyone in sync.
They built a Claude Code helper that suggests dates, nudges them after gaps, and may later handle reservations.
A reviewer tried Comet on shopping records, old email, photo sorting, and calendar setup, but the work was slow and often failed.
They attached an MCP server to Wiki.js so an LLM could edit household pages.
One Comet user had an agent send roughly 100 LinkedIn applications in 40 minutes, but none turned into interviews.
Uprent’s agent monitors Dutch listings, sends viewing requests, and helps manage landlord messages and calendars.
An agent searched Mubi for lesser-known films that fit a user’s preferences and added 10 picks to their watchlist.
A commenter had ChatGPT Agent enter the non-sensitive parts of a clinic refill form while they filled in the private fields by hand.