home / skills / openclaw / skills / reminder-research

reminder-research skill

/skills/nonlinear/reminder-research

This skill processes Apple Reminders automatically, applies list-based or custom research paths, and updates notes with clearly labeled results.

npx playbooks add skill openclaw/skills --skill reminder-research

Review the files below or copy the command above to add this skill to your agents.

Files (5)
SKILL.md
2.9 KB
---
name: reminder-research
description: "Natural language task queue via Apple Reminders. Agent executor: use skills (i-ching, librarian), edit files (ROADMAP, calendar), call APIs (GitHub, HA). Result tracking with šŸ¤– signifier. Triggers: reminders with notes (no šŸ¤–), heartbeat automated processing."
type: public
version: 2.0.0
status: published
dependencies:
  - remindctl
  - jq
requires:
  apis:
    - searxng (self-hosted, optional for web research)
  binaries:
    - remindctl (brew install steipete/tap/remindctl)
    - jq (brew install jq)
notes:
  - Requires macOS (Apple Reminders)
  - Cron scheduling recommended (3AM daily via LaunchAgent)
  - Agent can use any OpenClaw skill + tools
author: nonlinear
license: MIT
---

# Reminder Research

**Turn Apple Reminders into a natural language task queue**

**Published:** https://clawhub.ai/nonlinear/reminder-research

1. **Reminder has no notes?** → SKIP
2. **Reminder has notes?** → EXECUTE
3. **Reminder has notes with šŸ¤–?** → SKIP

**Ask questions. Trigger skills. Delegate tasks.**

---

## Installation

1. Install dependencies 
   `brew install steipete/tap/remindctl jq`

2. Authorize reminders access

   `remindctl authorize`

3. Install skill
   `clawd whatever put code here`

1. Setup cronjob

---

## How It Works

```mermaid
graph TD
    A[3AM Cron] -->|scans| B{Reminder}
    
    B -->|no notes| C[SKIP]
    B -->|has šŸ¤–| C
    B -->|notes, no šŸ¤–| D[Spawn Agent]
    
    D -->|executes| E[Skills/APIs/Files]
    E -->|updates| F[šŸ¤– Result]
```

---

## What You Need

- macOS + Apple Reminders
- `remindctl`: `brew install steipete/tap/remindctl`
- `jq`: `brew install jq`
- OpenClaw gateway running
- **Cron job** (LaunchAgent, 3AM daily)

---

## What It Does

Agent executes natural language commands:

āœ… **Research** (web, books, skills)  
āœ… **File operations** (edit ROADMAP, create notes, git commits)  
āœ… **Calendar** (create events, recurring schedules)  
āœ… **APIs** (GitHub issues, Home Assistant, Jira)  
āœ… **Automation** (anything you can describe)

**Natural language → Agent figures out HOW → Executes → Reports back with šŸ¤–**

---

## Examples

**Skills:**
```
Notes: "search iching hexagram 30 for love"
→ šŸ¤– Hexagram 30 (離 Li): Love requires clarity and passion...
```

**Roadmap:**
```
Notes: "add to personal roadmap: v0.9.0 - Calendar Control Plane"
→ šŸ¤– Added epic v0.9.0. Commit: a3f82b1
```

**Calendar:**
```
Notes: "create event Friday 3pm: Design review with Nicholas"
→ šŸ¤– Event created: Friday Feb 28 at 3:00 PM
```

**GitHub:**
```
Notes: "create issue in librarian repo: --book flag not working"
→ šŸ¤– Issue #47 created: https://github.com/.../issues/47
```

**Home Assistant:**
```
Notes: "turn off bedroom lights at 11pm daily"
→ šŸ¤– Automation created: automation.bedroom_lights_off
```

**Research:**
```
Notes: "web search: best iPad mini 6 deals under $350"
→ šŸ¤– FOUND: eBay $320, Swappa $340, Facebook $300
```

Overview

This skill processes Apple Reminders and turns empty or instruction-filled reminder notes into concise, tracked research results. It auto-detects reminders needing work, runs list-based defaults (system analysis, shopping, generic how-tos) or follows custom instructions that combine book and web sources. Results are written back to the reminder notes prefixed with the šŸ’Ž signifier for easy tracking. Heartbeat-driven scheduling keeps processing lean and cost-effective.

How this skill works

The skill queries Reminders via remindctl, filters incomplete items, and skips any note already starting with šŸ’Ž. Empty-note reminders are handled by list-based defaults: claw → memory/system analysis, Shopping → product/price research, others → generic web tutorials. Notes containing custom instructions trigger multi-source research (book librarian + web search) and obey explicit constraints before updating the note with a formatted šŸ’Ž research result.

When to use it

  • Automate product price and availability checks for Shopping list items.
  • Capture system/postmortem analysis for items in the claw list without manual lookup.
  • Run deep, constrained research that references specific books plus web sources.
  • Turn unattended todo reminders into short how-to summaries or links.
  • Schedule regular heartbeat runs to keep backlog researched and annotated.

Best practices

  • Leave custom instructions in the reminder note when you need Gen 3 multi-source research.
  • Use clear constraints (e.g., 'web search only', 'budget under $50', or specific book titles).
  • Avoid using the Groceries list for processing; groceries are skipped by design.
  • Grant remindctl permission and install jq to ensure reliable detection and editing.
  • Mark already-processed items by leaving the šŸ’Ž prefix to prevent duplicate work.

Example use cases

  • List: claw — Title: 'Messages disappear after reindexing' → Auto: memory_search + pattern analysis + actionable fixes in notes.
  • List: Shopping — Title: 'iPad mini 6, second hand' → Auto: web price sweep (eBay/Swappa) and summarized links/prices.
  • List: TODO — Title: 'Bitcoin regulation impact' with note requesting a specific book + SEC ruling → Custom: librarian + web_search and a combined summary.
  • List: Creative Code — Title: 'Vertical slider library' with constraints (MIT license, Framer Motion) → Custom: targeted GitHub code search and result links.

FAQ

How does the skill avoid repeated processing?

Processed reminders are marked by a leading šŸ’Ž in notes; the script skips any note already starting with that signifier.

Can I force web-only or book-only research?

Yes. Put explicit constraints in the notes (e.g., 'web search only' or 'search book X') and the skill will follow them when performing Gen 3 custom research.