A zero-trust structural database scanner that audits schema topology for performance bottlenecks and integrity gaps without accessing row data.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install graphcare
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 graphcare 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 GraphCare?
GraphCare serves as an essential structural antivirus for modern databases, specifically designed to handle the rapid schema evolution common in AI-driven development. By analyzing schema topology rather than row content, this addition to the Openclaw Skills library identifies hidden architectural flaws like circular dependencies and missing indexes that degrade performance over time.
Built with a zero-trust architecture, GraphCare ensures data privacy by strictly querying metadata catalogs such as information_schema and pg_indexes. It provides developers with a clear health score and actionable insights, making it a critical tool for maintaining high-performance PostgreSQL, MySQL, and SQLite environments without risking data exposure.
GraphCare Use Cases
- Auditing AI-generated database schemas to ensure structural integrity and best practices.
- Identifying the root cause of slow JOIN and DELETE operations caused by missing foreign key indexes.
- Detecting orphaned tables that are no longer reachable or integrated into the application logic.
- Preventing replication failures and ORM errors by identifying tables lacking primary keys.
- Optimizing write speeds and disk usage by locating redundant or duplicate indexes.
How GraphCare Works
- The user provides a database connection string to the audit_db_structure tool within the Openclaw Skills framework.
- GraphCare establishes a read-only connection to the target database's metadata catalog.
- The engine performs a topological scan of tables, foreign keys, and index definitions without executing any SELECT statements on user data.
- Identified issues are categorized by severity (Critical, Warning, Info) and compiled into a structured JSON report.
- Users can optionally call the explain_finding tool to receive plain-language remediation steps and impact analysis for specific structural issues.
GraphCare Setup
To integrate GraphCare into your environment, add it to your MCP client configuration:
{
"mcpServers": {
"graphcare": {
"command": "node",
"args": ["/path/to/graphcare/index.js"]
}
}
}
Alternatively, install the package globally via npm to use it as part of your Openclaw Skills workflow:
npm install -g graphcare-mcp
graphcare-mcp
For containerized environments, use Docker:
docker build -t graphcare .
docker run -i graphcare
GraphCare Data Schema & Taxonomy
GraphCare generates a comprehensive JSON health report following each scan. The data is organized into four primary sections:
| Key | Description |
|---|---|
| db_type | The detected database engine (PostgreSQL, MySQL, or SQLite). |
| tables[] | An array of all table names identified during the metadata crawl. |
| findings[] | A list of objects containing the issue type, severity level, affected table, and a descriptive message. |
| metrics{} | Aggregated counts of finding types and a calculated health_score (0-100). |
GraphCare Advanced Features
- Multi-database support across PostgreSQL, MySQL, and SQLite using a unified metadata extraction layer.
- Zero-Trust metadata scanning that guarantees no user row data is ever read or transmitted.
- Automated health scoring to provide a high-level metric for CI/CD pipeline integration.
- Context-aware remediation advice through the explain_finding tool to assist developers in rapid debt reduction.
- Stateless execution model ensuring memory is purged after every scan for maximum security within Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
Coding Style Memory
A local, consent-based memory skill that allows AI agents to adapt to your specific coding styles and stack preferences through explicit feedback.
Book Highlights
A specialized AI agent skill designed to extract and format the most impactful highlights, quotes, and core insights from an extensive library of books.
CrewAI
CrewAI is a sophisticated multi-agent orchestration framework designed to help developers build collaborative AI systems where specialized agents work together to solve complex tasks.
Zhuihui Branch Attendance Analyzer
An automated analysis tool for processing complex employee attendance records and generating professional docx and xlsx reports.
Knowledge Capture
An intelligent knowledge management and adaptive review system that converts conversations into structured Markdown notes and automated study plans.
PPTMagic PPT Creator
A powerful AI skill for generating, continuing, and exporting professional presentation decks via the pptmagic and Banana Slides API ecosystem.