Zhixin Graph is a dynamic knowledge-trust graph system that combines atomic knowledge facts with a verifiable trust chain for AI agents and human collaborators.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install zhixin-graph
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 zhixin-graph 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 Zhixin Graph (Knowledge-Trust Graph)?
Zhixin Graph is a sophisticated tool designed to bridge the gap between raw information and trusted knowledge. It operates on a three-layer model consisting of entities, relations, and facts, uniquely integrated with a dynamic verification chain. Unlike static knowledge bases, this skill treats trust as an evolving attribute, allowing users to track who provided information, how it was verified, and whether it remains valid over time. It is a core component of the Openclaw Skills ecosystem, designed for seamless integration with AI agents like Claude Code and OpenClaw.
The system is built with a philosophy of human-centric AI assistance, where the AI helps organize, relate, and flag verification needs while humans retain final decision-making authority. With zero external dependencies beyond the Python standard library and optional Ollama support for semantic search, it offers a high-performance, private-by-default solution for personal knowledge management and small team collaboration.
Zhixin Graph (Knowledge-Trust Graph) Use Cases
- Organizing fragmented information from meetings or documents into a structured knowledge graph.
- Tracking the verification status of project decisions and technical specifications.
- Managing team-wide shared intelligence with conflict resolution and peer review cycles.
- Performing semantic or keyword-based searches across a local repository of verified facts.
- Auditing the provenance of information to ensure data reliability in AI-assisted workflows.
How Zhixin Graph (Knowledge-Trust Graph) Works
- Extraction: The user or an AI agent extracts entities (people, projects, concepts), relations (works_on, depends_on), and facts from a source.
- Initialization: Knowledge is entered into the system via CLI, defaulting to an unverified state.
- Verification: A verification chain is established, moving facts from unverified to self-checked, peer-reviewed, or consensus status.
- Evolution: The graph updates dynamically as new annotations, disputes, or confirmations are added by team members.
- Maintenance: Stale information is automatically flagged based on expiry dates (fresh_until), prompting updates or re-verification.
Zhixin Graph (Knowledge-Trust Graph) Setup
Zhixin Graph requires Python 3.8+ and minimal setup. It stores all data locally in a SQLite database.
# Initialize the knowledge base
python3 ~/.zhixin/kg.py init
# Check the status of the database
python3 ~/.zhixin/kg.py stats
# (Optional) Generate embeddings for semantic search if Ollama is installed
python3 ~/.zhixin/kg.py embed
Zhixin Graph (Knowledge-Trust Graph) Data Schema & Taxonomy
The system organizes data using a relational schema optimized for graph traversal and trust auditing:
| Level | Component | Description |
|---|---|---|
| Knowledge Layer | Entities | Nodes representing people, tools, concepts, or projects. |
| Relations | Directed edges defining how entities interact (e.g., A depends_on B). | |
| Facts | Atomic statements attached to entities with specific types (knowledge, decision, etc.). | |
| Trust Layer | Sources | Metadata on where information originated. |
| Verification Chain | Tracking the level of trust: unverified to external_confirmed. | |
| Annotations | Team feedback, disputes, and confirmations for each fact. | |
| Context Layer | Branches | Isolated units for different projects or teams. |
Zhixin Graph (Knowledge-Trust Graph) Advanced Features
- Multi-branch management for isolating knowledge across different projects or teams.
- Semantic search capabilities powered by local Ollama embedding models.
- Automated stale-data reporting to maintain knowledge freshness and accuracy.
- AI-assisted batch extraction workflow for rapid knowledge ingestion from raw text.
- JSON-based export/import for seamless team synchronization without central servers.
SKILL.md
Loading
Related Openclaw Skills
VPick AI Video
A cloud-powered AI video editing skill that automatically selects and compiles the best moments from raw footage into high-quality highlight clips.
Tomoviee Video Background
An AI-powered video editing skill that replaces backgrounds and generates 1080p content through simple natural language commands.
Blender Render
A headless automation skill for generating high-quality technical and beauty renders of 3D assets using Blender.
Editor Maker
A cloud-based AI video editing skill that automates the process of cutting, styling, and exporting raw video clips into professional 1080p MP4 files.
Address Change Mailing Label Checklist
A comprehensive printable organizer to manage address updates across all service providers, government agencies, and personal contacts during a move.
Web Monitor
An automated web monitoring tool that tracks page changes using CSS selectors and hash-based comparison.
Featured*
