rootcause-mcp
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@rootcause-mcpPerform root cause analysis on a fall incident using fishbone diagram"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
RootCause MCP - Clinical Causality & Teaching Design MCP Server
π₯ AI-guided causal analysis, feedback-loop modeling, and teaching-case construction for healthcare quality improvement
English | δΈζη
π― Vision
Enable AI Agents to move in both directions of causality:
from clinical incidents β structured, traceable root cause analysis
from root causes β learner-ready lesson plans for medical education
RootCause MCP supports three categories of analysis models through Domain Cartridges:
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β RootCause MCP β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ€
β βββββββββββββββ βββββββββββββββ βββββββββββββββ β
β β PROSPECTIVE β βRETROSPECTIVEβ β SYSTEMIC β β
β β Proactive β βInvestigationβ β Complexity β β
β βββββββββββββββ€ βββββββββββββββ€ βββββββββββββββ€ β
β β β’ HFMEA β β β’ HFACS β
β β β’ STAMP/STPAβ β
β β β’ HVA β β β’ 5-Whys β
β β β’ FRAM β β
β β β’ Bowtie β β β’ Fishboneβ
β β β’ AcciMap β β
β βββββββββββββββ βββββββββββββββ βββββββββββββββ β
β βΌ β
β βββββββββββββββββββββββββ β
β β Unified Graph API β β
β β (21 MCP Tools) β β
β βββββββββββββββββββββββββ β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββRelated MCP server: MedVision MCP
β¨ Core Features
Retrospective Cartridge (Implemented β )
Feature | Description | Status |
π Fishbone (6M) | Healthcare-specialized Ishikawa diagram | β 4 tools |
π 5-Why Analysis | Deep cause exploration with Proximate/Ultimate classification | β 4 tools |
π Bidirectional Causality | Cross-links, escalation loops, and feedback cycle modeling | β 1 tool |
π Teaching Case Builder | Convert RCA chains into medical student lesson plans | β 1 tool |
π HFACS-MES | Human Factors Analysis auto-suggestion (5-level, 25 categories) | β 6 tools |
β Causation Verify | Bradford Hill criteria-based verification | β 1 tool |
π 6M-HFACS Mapping | Cross-reference between taxonomies | β 1 tool |
πΎ Session Management | Persistent analysis sessions | β 4 tools |
Prospective Cartridge (Planned π)
HFMEA - Healthcare Failure Mode and Effect Analysis
HVA - Hazard Vulnerability Analysis
Bowtie - Threat and consequence analysis
Systemic Cartridge (Planned π)
STAMP/STPA - Control loop analysis
FRAM - Functional Resonance Analysis Method
π§ Available Tools
HFACS Tools (6)
Tool | Description |
| Auto-suggest HFACS codes from cause description |
| Confirm or override HFACS classification |
| Get full HFACS-MES framework structure |
| List learned classification rules |
| Hot-reload YAML rules |
| Get 6M-HFACS cross-reference table |
Session Tools (4)
Tool | Description |
| Create new RCA session |
| Get session details |
| List all sessions |
| Archive completed session |
Fishbone Tools (4)
Tool | Description |
| Initialize fishbone diagram |
| Add cause to 6M category |
| Get fishbone structure |
| Export as Mermaid/Markdown/JSON |
Why Tree Tools (6)
Tool | Description |
| Progressive 5-Why questioning |
| Get Why tree structure |
| Mark node as root cause |
| Export as Mermaid/Markdown/JSON |
| Add feedback loops and bidirectional causal links |
| Build a teaching-ready lesson plan from an RCA chain |
Verification Tools (1)
Tool | Description |
| Verify causation with 4-criteria test |
π Quick Start
# Install with uv (recommended)
uv pip install -e .
# Or development mode
uv pip install -e ".[dev]"
# Run MCP Server
python -m rootcause_mcp.interface.serverVS Code Integration
Add to your .vscode/mcp.json:
{
"servers": {
"rootcause-mcp": {
"type": "stdio",
"command": "uv",
"args": ["run", "python", "-m", "rootcause_mcp.interface.server"],
"cwd": "${workspaceFolder}"
}
}
}π Project Structure
rootcause-mcp/
βββ src/rootcause_mcp/
β βββ domain/ # Domain Layer (DDD)
β β βββ entities/ # Session, Cause, Fishbone, WhyNode
β β βββ value_objects/# HFACSCode, Scores, Identifiers
β β βββ repositories/ # Repository interfaces
β β βββ services/ # HFACSSuggester, CausationValidator
β βββ application/ # Application Layer
β β βββ session_progress_tracker.py
β β βββ guided_response_builder.py
β βββ infrastructure/ # Infrastructure Layer
β β βββ persistence/ # SQLite + SQLModel
β βββ interface/ # Interface Layer
β βββ tools/ # MCP Tool definitions
β βββ handlers/ # Tool handlers
β βββ server.py # MCP Server entry
βββ config/hfacs/ # YAML configurations
β βββ hfacs_mes.yaml # HFACS-MES framework
β βββ fishbone_6m.yaml # Healthcare 6M categories
β βββ keyword_rules.yaml# Classification rules
βββ tests/ # Test suites
βββ docs/ # Documentation
β βββ spec_v2.md # Full specification v2.5.0
βββ memory-bank/ # Project memory (for AI)π Related MCPs
MCP | Purpose | Integration |
Data decomposition & table refactoring | Pre-processing | |
HHRAG MCP | Knowledge graph retrieval | Context enrichment |
CGU MCP | Creative divergent thinking | Cause brainstorming |
π Documentation
Full Specification - v2.5.0 (3700+ lines)
Architecture - DDD layers
Roadmap - Cartridge expansion plan
Literature Review - HFACS-MES research
π§ Development
# Run tests
pytest tests/ -v
# Code linting
ruff check src tests
mypy src --strict
# Security scan
bandit -r src
# Dead code detection
vulture srcπ€ Contributing
Contributions welcome! Please see CONTRIBUTING.md.
π License
Apache 2.0 License - See LICENSE
Made with β€οΈ for Healthcare Quality Improvement
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Latest Blog Posts
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/u9401066/rootcause-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server