mcp-factcheck
A Model Context Protocoll (MCP) server that validates content against official MCP specification using semantic search and AI-powered validation.
Install / Use
claude mcp add carlisia -- npx -y github:carlisia/mcp-factcheckIf the server publishes to npm under a different name, use that package instead — check the repo README.
MCP Server
Model Context Protocol server
Quality Score
Category
Development & EngineeringSupported Platforms
Tags
Our assessment of mcp-factcheck
mcp-factcheck scores 77/100 on our quality scale, 352nd of 1,283 Development & Engineering skills we index (top 28%).
Its MCP Server is 7.1 KB long, well organised into 25 sections with 7 code examples: a thorough specification that gives an agent plenty to work with.
It has 3 GitHub stars, so there is little community track record yet; judge it on its content.
Maintenance, license and trust
- The repository was last updated about 10 months ago. That is recent enough to be usable, but agent tooling moves fast, so check the instructions against your agent's current version.
- It is released under the MIT license, a permissive license that allows use, modification and commercial use with attribution.
- Its trust signals score 86/100, with 2 cautions from licensing, adoption, age or documentation. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
mcp-factcheck compared with similar skills
All 4 of these similar skills score higher than mcp-factcheck; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| mcp-factcheck (this skill)by carlisia | 77 | 3 | 10mo ago | MCP Server |
| Agent-Reachby Panniantong | 100 | 85.0k | 8d ago | CLAUDE.md |
| headroomby headroomlabs-ai | 100 | 73.6k | today | CLAUDE.md |
| rufloby ruvnet | 100 | 73.1k | today | CLAUDE.md |
| career-opsby career-ops-hq | 100 | 72.5k | today | CLAUDE.md |
Frequently asked questions
- How do I install mcp-factcheck?
- Run
claude mcp add carlisia -- npx -y github:carlisia/mcp-factcheck. The install tabs above show the steps for each supported agent. - Which AI agents does mcp-factcheck work with?
- It is written for Claude Code and Claude Desktop, as a MCP Server file. Other agents that read the same format can often use it too.
- Is mcp-factcheck safe to use?
- It is MIT-licensed and scores 86/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is mcp-factcheck still maintained?
- The repository was last updated about 10 months ago. That is recent enough to be usable, but agent tooling moves fast, so check the instructions against your agent's current version.
Skill content
View source on GitHubMCP Fact-Check MCP Server
An MCP Server for validating code or content against the official Model Context Protocol (MCP) specification to ensure technical accuracy and prevent the spread of misinformation.
📦 View in MCP Registry - Available in the official MCP Registry
📋 View Project Roadmap - See planned features and development progress
🏗️ Design Documentation - Technical design and implementation details
Overview
The MCP Fact-Check MCP Server helps ensure technical accuracy when coding or writing about MCP by comparing content against official specifications. It uses:
- Semantic search with OpenAI embeddings to find relevant specification sections
- AI-powered validation to detect inaccuracies and suggest corrections
- Compound claim decomposition to validate complex statements with multiple assertions
- Multiple spec versions support (draft, 2025-06-18, 2025-03-26, 2024-11-05)
Features
MCP Tools Exposed
-
check_mcp_claim- Comprehensive validation of MCP-related content- Validates multi-claim content (documentation, tutorials, bullet points)
- Automatically decomposes compound claims (e.g., "X and Y") for accurate validation
- Provides step-by-step validation workflow
- Identifies missing best practices and modal verb issues
- Returns corrected content with confidence scores
-
check_mcp_quick_fact- Quick fact-checking for single MCP claims- Validates single sentences or quick questions
- Returns concise ✓/✗ verdict with explanation
- Uses aggressive search strategies for accuracy
- Perfect for "Does MCP support X?" questions
-
search_spec- Searches MCP specifications using semantic similarity- Returns most relevant specification sections
- Supports all specification versions
-
list_spec_versions- Lists available MCP specification versions- Shows version dates and descriptions
- Indicates which version is current
MCP Prompts Available
-
migrate-mcp-content- Guides content migration between MCP specification versions- Validates content against source specification first
- Identifies changes between specification versions
- Provides step-by-step migration guidance
- Works with any type of MCP-related content
- Preserves the original tone, style, and voice when making corrections or suggestions
Parameters:
current_version(required): Source MCP specification version (e.g., "2024-11-05", "2025-06-18")target_version(required): Target MCP specification version to migrate to (e.g., "draft")update_scope(optional): Determines how aggressive the migration should becritical_only: Fix only critical inaccuracies and breaking changes (minimal changes)enhancement_focused: Fix issues and improve clarity, align with best practicescomprehensive: Complete review with all improvements and enhanced clarity- Default:
comprehensive
Installation
The MCP Fact-Check server is available through the Model Context Protocol registry. Install it directly from your MCP client:
For Claude Desktop and other MCP clients:
- Search for "mcp-factcheck" in your client's server marketplace
- Click install
- Provide your OpenAI API key when prompted
That's it! The server will be automatically configured and ready to use.
For developers: If you need to build from source or contribute to the project, see INSTALL.md for development setup instructions.
Observability
Visual Tracing with Arize Phoenix
For a beautiful, AI-focused trace visualization UI, set up Arize Phoenix:
- Install and start Phoenix:
# Install Phoenix
pipx install arize-phoenix
# Start Phoenix server
phoenix serve
- Update the Host config to send traces to Phoenix:
{
"mcpServers": {
"mcp-factcheck": {
"command": "/path/to/bin/mcp-factcheck-server",
"args": [
"--data-dir",
"/path/to/data/embeddings",
"--telemetry",
"--otlp-endpoint",
"http://localhost:6006"
],
"env": {
"OPENAI_API_KEY": "your-api-key"
}
}
}
}
- View traces at: http://localhost:6006
What you'll see in Phoenix:
- Beautiful AI-focused interface designed for LLM applications
- Complete validation pipeline timeline with clear visual hierarchy
- Embedding generation performance and OpenAI API call tracking
- Vector search visualization with similarity scores
- Per-chunk validation confidence levels and quality metrics
- Cost tracking for OpenAI API usage (or whichever llm is being used for embedding the input content/code)
- Clean, intuitive navigation focused on AI workflows
Phoenix is specifically designed for AI/ML observability and provides a much more user-friendly experience than traditional tracing tools.
Development
Building
# Build all components
go build -o bin/mfc ./cmd/server
go build -o bin/specloader ./utils/cmd
# Run tests
go test ./...
Updating Specifications
The project includes pre-extracted MCP specifications and embeddings for all versions. To check when the draft specification was last updated, see data/SPEC_METADATA.json:
# View draft update information
cat data/SPEC_METADATA.json | jq '.specs.draft'
To update the draft specification:
./bin/specloader spec --version draft
./bin/specloader embed --version draft
./bin/specloader embed --version draft-fine
To add a new specification version:
./bin/specloader spec --version 2025-12-15
./bin/specloader embed --version 2025-12-15
./bin/specloader embed --version 2025-12-15-fine
All specification extraction dates and source commits are automatically tracked in data/SPEC_METADATA.json.
Testing Tools
Test the server using the included test client:
# Build test client
go build -o bin/factcheck-curl ./cmd/factcheck-curl
# Test tools
./bin/factcheck-curl --cmd ./bin/mfc --data-dir ./data/embeddings tools/list
./bin/factcheck-curl --cmd ./bin/mfc --data-dir ./data/embeddings tools/call validate_content '{"content":"MCP is a protocol"}'
# Test prompts
./bin/factcheck-curl --cmd ./bin/mfc --data-dir ./data/embeddings prompts/list
# Get migration prompt with minimal parameters
./bin/factcheck-curl --cmd ./bin/mfc --data-dir ./data/embeddings prompts/get migrate-mcp-content '{"current_version":"2024-11-05","target_version":"draft"}'
# Get migration prompt with all parameters
./bin/factcheck-curl --cmd ./bin/mfc --data-dir ./data/embeddings prompts/get migrate-mcp-content '{
"current_version": "2024-11-05",
"target_version": "2025-06-18",
"update_scope": "critical_only"
}'
Architecture
See DESIGN.md for the complete architecture documentation.
Environment Variables
OPENAI_API_KEY- Required for embedding generation and content validationGITHUB_TOKEN- Optional, for higher GitHub API rate limits when extracting specs
License
MIT License. See LICENSE for details.
Related Skills
Agent-Reach
85.0kGive your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.
headroom
73.6kCompress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
ruflo
73.1k🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated
career-ops
72.5kOpen-source AI job search: scan job portals, evaluate listings into a structured A-H report with a global 1-5 score, tailor your CV, track applications — runs locally in your AI coding CLI (Claude Code, Codex, OpenCode, Antigravity…)
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
