Relation Keeper for Openclaw

An AI-powered social relationship manager that tracks personal profiles, archives shared experiences, and automates event reminders.

yujintang
v1.0.1
Feb 5, 2026
1
2.1k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install relation-keeper

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install relation-keeper using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Relation Keeper?

Relation Keeper is a specialized social management skill designed for AI agents to maintain and remember complex social networks. It serves as a personal CRM, enabling users to record character portraits including gender, birthdays, addresses, and unique facts like personal hobbies or professional roles. By integrating with Openclaw Skills, it ensures that your social interactions are never forgotten and your upcoming commitments are proactively managed.

The skill provides a structured way to store both the history of your relationships and the future of your social calendar. It bridges the gap between raw conversation and actionable data, allowing an AI agent to recall who someone is, what you have done together, and what you have planned next without manual data entry.

Relation Keeper Use Cases

  • Creating and updating detailed contact profiles from casual conversation snippets.
  • Setting multi-stage automated reminders for birthdays and anniversaries (7 days, 3 days, and day-of).
  • Archiving past shared experiences like dinners, meetings, or trips for future reference.
  • Quick retrieval of contact details such as phone numbers, addresses, or personal preferences.
  • Managing future appointments with lead-time notifications.

How Relation Keeper Works

  1. The AI agent identifies names, dates, and relationship facts within user input.
  2. Information is categorized into portrait data, past event archives, or future event schedules.
  3. The skill updates local JSON storage, merging new facts into existing profiles or creating new ones.
  4. A background cron job runs every 15 minutes to scan the database for upcoming events.
  5. Notifications are triggered based on specific rules, such as 2 hours before an appointment or multiple days before a birthday.
  6. Sent reminders are logged to ensure no duplicate notifications are delivered.

Relation Keeper Setup

Set the required environment variables in your shell or .env file to define data paths and timezones:

export RELATION_KEEPER_DATA="$HOME/.openclaw/relation-keeper"
export RELATION_KEEPER_TZ="Asia/Shanghai"

Install the skill dependencies and initialize the automated scanning task:

npm install
npm run install:cron

Relation Keeper Data Schema & Taxonomy

The skill maintains data integrity through a set of structured JSON files:

File Content Description
portraits.json Stores person IDs, aliases, gender, birth details, address, and an extensible list of facts.
past_events.json A chronological log of shared activities including type, date, and participants.
future_events.json Upcoming events tagged with reminder rules (birthday or appointment).
reminders_sent.json A tracking file to manage the state of delivered notifications.

Each portrait supports a facts array where the AI can append key-value pairs like 'Hobbies' or 'Job Title' with source tracking.

Relation Keeper Advanced Features

  • Multi-stage reminder logic: Birthday rules trigger at 7-day, 3-day, and 0-day intervals, while appointments trigger at 2-hour and 0-hour marks.
  • Natural Language Date Resolution: Supports processing complex time expressions like 'next Friday evening' into standardized ISO timestamps.
  • Configurable Notification Channels: Support for routing reminders to specific platforms like Telegram via the RELATION_KEEPER_CHANNEL variable.
  • Automated Cron Management: Self-configuring background tasks that remove the need for manual scheduling of scan scripts.
  • Contextual Recall: Ability to summarize relationship history by filtering past events based on participant IDs.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*