ERDL MCP Server
Enforces deterministic rules for AI agents with 30 built-in presets and unlimited personal rules to ensure consistent behavior.
README
ERDL MCP Server
Deterministic rules for AI Agents. Say it once. Never repeat.
ERDL (Entity-Rule Definition Language) gives your Agent deterministic rules. 30 built-in presets. Unlimited personal rules. Free forever.
Why ERDL?
| Without ERDL | With ERDL |
|---|---|
"Don't use any" — forgotten in 5 turns |
Rules enforced deterministically, every tool call |
| "Write shorter sentences" — ignored | Writing rules fire before every output |
| "Confirm before adding deps" — skipped | Blocked until you confirm |
| "Why did you do that?" — no answer | erdl_explain shows the full decision trail |
Quick Start
npx -y @openoba-ai/erdl-mcp
30 rules active immediately. No account. No configuration. No API key.
中文用户:
npx -y @openoba-ai/erdl-mcp --lang zh
Free vs Pro
| Free | Pro | |
|---|---|---|
| 5 MCP Tools | ✅ | ✅ |
| 30 Preset Rules | ✅ | ✅ |
| Unlimited Personal Rules | ✅ | ✅ |
| Chinese / English | ✅ | ✅ |
| All 11 Operators | ✅ | ✅ |
| MIT Open Source | ✅ | ✅ |
| Execution Rings 1–2 (REQUEST_HUMAN, ESCALATE, ROLLBACK, QUARANTINE) | — | ✅ |
| Guardian Agent Role (监管者模式) | — | ✅ |
| Audit Export (OCSF / OTLP) | — | ✅ |
| Team Rules (组织级规则共享) | — | ✅ |
| Dashboard (规则命中统计) | — | ✅ |
| Enterprise Compliance Packs (GB/Z 185, NIST, EU AI Act) | — | Enterprise |
One-command Install
# macOS / Linux
curl -fsSL https://raw.githubusercontent.com/OpenOBA/erdl-mcp-server/master/scripts/install.sh | bash
# Windows (PowerShell)
iwr https://raw.githubusercontent.com/OpenOBA/erdl-mcp-server/master/scripts/install.ps1 | iex
Why ERDL vs SKILL.md / Prompt Rules?
| Prompt-Based Rules | ERDL MCP Server | |
|---|---|---|
| Execution | LLM "tries" to follow | Deterministic engine — guaranteed |
| Visibility | Invisible, can't tell if it worked | erdl_explain shows every decision |
| Reliability | LLM may ignore or forget | Zero hallucination — condition match is mathematical |
| Testing | Manual | erdl_simulate with 3 auto-scenarios |
| Portability | Single platform | All MCP-compatible Agents |
ERDL rules don't suggest. They don't hope. They enforce.
What's Included
30 Built-in Rules
| Category | Rules | Scope |
|---|---|---|
coding |
10 | TypeScript quality, Git discipline, dependency hygiene |
engineering |
10 | Workflow discipline: honesty, no shortcuts, pipeline gates |
writing |
7 | Tone, formatting, clarity |
design |
3 | Tailwind-first, responsive, accessible |
5 MCP Tools
| Tool | Purpose |
|---|---|
erdl_evaluate |
Action Guard — check rules before every tool call |
erdl_simulate |
Test a rule against 3 scenarios before creating |
erdl_create_rule |
Create a rule from natural language |
erdl_list_rules |
List all active rules |
erdl_explain |
Full decision trail for any action |
MCP Resources
| Resource | Content |
|---|---|
erdl://rules/list |
All rules as JSON |
erdl://status |
Server runtime status |
Installation
Any MCP Client
{
"mcpServers": {
"erdl": {
"command": "npx",
"args": ["-y", "@openoba-ai/erdl-mcp@latest"]
}
}
}
Or let ERDL generate the config for your client:
npx @openoba-ai/erdl-mcp@latest --setup
Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or the equivalent path on your OS.
Cursor
Add to .cursor/mcp.json in your project root.
OpenClaw
openclaw mcp set erdl '{"command":"npx","args":["-y","@openoba-ai/erdl-mcp@latest"]}'
VS Code / GitHub Copilot
Add to .vscode/mcp.json or the Copilot MCP configuration.
Creating Your First Rule
You: "Never use `any` in TypeScript."
Agent: 1. erdl_simulate → tests the rule against 3 scenarios
2. erdl_create_rule → saves to ~/.openoba/rules/
3. Rule is now active. Next `any` will be blocked.
You: "Why did you reject my code?"
Agent: erdl_explain → shows every rule that fired
Or create rules manually in ~/.openoba/rules/:
rules:
- id: MY-001
name: no_console_log
description: Don't commit console.log statements
category: coding
triggers: [write_file, edit, apply_patch]
when: "tool.args.path match \"\\.tsx?$\""
then: DENY "Remove console.log before committing"
priority: 10
enabled: true
📚 Full tutorial: Create Your First ERDL Rule → (English)
CLI Reference
npx @openoba-ai/erdl-mcp@latest # Start MCP server (always latest)
npx @openoba-ai/erdl-mcp@latest --lang zh # Chinese mode
npx @openoba-ai/erdl-mcp@latest --pro-key sk-xxx # Activate Pro license
npx @openoba-ai/erdl-mcp@latest --upgrade # Upgrade to latest
npx @openoba-ai/erdl-mcp@latest --uninstall # Clean removal
npx @openoba-ai/erdl-mcp@latest --help # Full usage
Stuck on an old version? npx caches your first install. If
--versionshows an older release, clean the cache:npm cache clean --forceThen re-run:
npx @openoba-ai/erdl-mcp@latest
Run from Source
git clone https://github.com/OpenOBA/erdl-mcp-server.git
cd erdl-mcp-server
npm install
npm run build
node bin/erdl-mcp.js --version
License
Built by AI + Human. Deterministic architecture, not prompt engineering.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.