Vai al contenuto
TakoAPI
Torna alle skill

waste-reminder

di apenklit

Productivity & Tasks
471Download riportati da clawskills.sh
0Visualizzazioni
0 likeAccedi per mettere like a questa skill
—Non ancora valutato
La tua valutazione

Sintesi

A flexible, token-efficient skill for automated waste container collection reminders.

Descrizione

Cosa fa questa skill

Automates waste collection reminders based on user-defined pickup schedules and container types. Reads a JSON config, checks which reminders are due, and outputs formatted messages for the AI to send to configured channels. Supports up to 4 reminder times per pickup with group, personal, and escalation targeting.

Schedule logic runs outside the AI context so each 15-minute cron check consumes no AI tokens, keeping costs low for frequent polling.

Quando usarla

  • Alerting a household WhatsApp group the evening before bin collection day
  • Sending an escalation reminder to a specific person if no one confirms the container is out
  • Managing separate schedules for paper, plastic, garden, and residual waste containers
  • Routing morning reminders to a partner on a different channel than group notifications
  • Stopping repeat reminders automatically once a container is confirmed as put out

Esempio di workflow

INPUT

User asks: set up waste reminders for paper and plastic containers with pickups on March 15 and March 16

AGENT
  1. 1

    Creates config.json with container definitions and reminder times at 18:00, 22:00, and 06:30

  2. 2

    Creates schedule.json with both pickup dates and all confirmation flags set to false

  3. 3

    Adds a cron job that runs waste_cron.py every 15 minutes

  4. 4

    At 18:00 on March 14, the cron script outputs a SEND_TO block targeting the group WhatsApp ID

  5. 5

    AI reads the output and delivers the message to the configured WhatsApp group

OUTPUT

Household receives: "Tomorrow: 🔵 Paper will be collected!"

Installazione

clawhub install apenklit/waste-reminder

Esegui il comando CLI oppure incolla il prompt in Claude Code, Cursor, Codex o nell'agente di coding che usi.

Link