A centralized intelligence hub that orchestrates autonomous relationship matching between individual user agents in a privacy-preserving manner.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install social-hub-server
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install social-hub-server 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 Matching Engine?
The Matching Engine serves as the central brain of a decentralized relationship system. Unlike standard agents, it functions as an independent instance within the Openclaw Skills framework, communicating exclusively with other personal agents rather than human users. It acts as a trusted intermediary, processing user profiles and metadata to identify high-value interpersonal connections.
By leveraging advanced LLM analysis, the engine evaluates complementarity and situational consistency to move beyond simple keyword matching. It handles the full lifecycle of a connection—from initial discovery and vector-based screening to two-way confirmation and the generation of personalized icebreakers—all while strictly adhering to user-defined privacy settings.
Matching Engine Use Cases
- Automated professional networking within private organizations or communities.
- Dynamic mentorship pairing based on evolving skill sets and goals.
- Project collaborator discovery using capability complementarity analysis.
- Privacy-preserving social discovery where agents negotiate data disclosure.
How Matching Engine Works
- Heartbeat Monitoring: The engine maintains a global registry by tracking status updates and last-seen timestamps from all connected personal agents.
- Profile Vectorization: When a user profile is updated, the engine converts tags into embeddings and updates relevant ChromaDB collections.
- Multi-Stage Evaluation: Potential matches undergo a three-tier process: frequency limit checks, vector similarity screening, and deep LLM compatibility scoring.
- Secure Negotiation: The engine sends match notifications and manages the two-way acceptance flow (MATCH_ACCEPT/MATCH_REJECT) between agents.
- Connection Finalization: Once both parties accept, the engine generates icebreakers and reveals shared data based on granular disclosure permissions.
Matching Engine Setup
To deploy the Matching Engine within your Openclaw Skills environment, ensure ChromaDB is installed and configure the local storage directories as follows:
# Create the engine data directory
mkdir -p ~/.matchbot-engine/chromadb/
# Initialize the registry and match history files
touch ~/.matchbot-engine/registry.json
touch ~/.matchbot-engine/match_history.json
Configure the engine to listen to the internal group communication channel for PROFILE_UPDATE and HEARTBEAT events to begin the orchestration process.
Matching Engine Data Schema & Taxonomy
The system organizes data into three primary layers for efficient retrieval and matching:
| Component | Path | Description |
|---|---|---|
| Global Registry | registry.json |
Stores user status, agent IDs, profile versions, and privacy disclosure settings. |
| Match History | match_history.json |
Tracks match IDs, compatibility scores (consistency/complementarity), and confirmation status. |
| Vector Store | /chromadb/ |
Contains specialized collections for skills, interests, goals, and challenges. |
ChromaDB Collections:
all_skills: Vectorized user capabilities for competency-based pairing.all_goals: Short and long-term user objectives.all_challenges: Current pain points used to trigger complementarity-driven matches.
Matching Engine Advanced Features
- Event-Driven Triggering: Real-time matching calculations triggered immediately upon profile updates to ensure high-velocity networking.
- Scheduled Global Scans: Periodic cron-based scanning (e.g., every 6 hours) to re-evaluate the entire user pool and catch missed opportunities.
- Disclosure-Aware Data Filtering: Advanced logic that only reveals sensitive profile fields after mutual acceptance, respecting user privacy.
- Observability Logging: Real-time system status reports sent to management groups, summarizing online users, active scans, and successful match rates.
SKILL.md
Loading
Related Openclaw Skills
Ads Update
An automated monitoring tool that tracks specific YouTube and Vimeo channels for new video releases and generates detailed performance reports.
Find Reference Video
Automate the search, selection, and localized downloading of film and commercial video references from frameset.app.
AI Twitter Digest
An automated monitoring tool that fetches, summarizes, and delivers daily AI and tech insights from Twitter to Discord or other messaging platforms.
AI News Feed
A real-time AI and ML news aggregator that provides LLM-enriched summaries and importance scores from over 57 curated industry sources.
Personal Relationship Agent
A local-first AI agent that synthesizes user profiles through natural dialogue to deliver intelligent relationship matching and networking opportunities.
Zillow Airbnb Matcher
A sophisticated real estate analysis tool that identifies properties for sale with proven short-term rental income potential by cross-referencing Zillow and Airbnb data.
Featured*
