MCP Tool Guardian
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., "@MCP Tool GuardianValidate the output from getUsers against the user schema"
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.
MCP Tool Guardian
Production-ready MCP Server for AI Agent Reliability & Accountability
Protect your AI agents from bad tool outputs, schema mismatches, and potential tool poisoning. Built for entrepreneurs who want safer, more accountable AI systems.
The Problem (Simple Explanation)
Imagine your AI helper is like a 10-year-old kid who asks friends for information. Sometimes the friends give wrong, incomplete, or even tricky answers. The kid then makes mistakes because of that bad information.
MCP Tool Guardian is like a smart teacher who checks every answer the friends give before the kid uses it. It makes sure the answer matches the expected shape, cleans dangerous parts, and gives a reliability score.
This is especially useful for founders building AI products who care about trust and accountability (like Prevalid's mission).
Related MCP server: Prompt Rejector MCP
What This MCP Server Does
It exposes tools that any MCP-compatible AI agent (Claude Desktop, Cursor, Windsurf, etc.) can call:
Tool | What it does (kid-friendly) |
| Checks if the data looks exactly like the promised shape (like checking if a LEGO set has all the right pieces) |
| Cleans the data – removes secrets, weird characters, or possible injection tricks |
| Gives a score from 0-100 how trustworthy the tool answer is |
| Looks for common "poison" patterns that bad tools sometimes hide |
| Writes a simple log so you can later see what happened (accountability!) |
Why This is Valuable for Entrepreneurs
Reduce AI failures: Fewer cascading errors from bad tool data
Low cost: Pure local computation – no expensive external APIs needed
Accountability: Perfect for Prevalid-style AI Execution OS
Ready for production: Stateless design compatible with new MCP 2026-07-28 spec
Easy to sell: Can be packaged as a paid MCP or internal reliability layer
Quick Start
Requirements
Python 3.10+
mcppackage (official SDK)
pip install mcp jsonschema pydanticRun the server
python -m src.serverOr with uv:
uv run python -m src.serverAdd to Claude Desktop / Cursor
Add this to your MCP config:
{
"mcpServers": {
"tool-guardian": {
"command": "python",
"args": ["-m", "src.server"],
"cwd": "/path/to/mcp-tool-guardian-20260730"
}
}
}Project Structure
mcp-tool-guardian-20260730/
├── src/
│ ├── __init__.py
│ ├── server.py # Main MCP server
│ ├── tools/
│ │ ├── __init__.py
│ │ ├── validate.py
│ │ ├── sanitize.py
│ │ ├── reliability.py
│ │ └── audit.py
│ └── utils/
│ └── schemas.py
├── tests/
│ └── test_tools.py
├── mcpize.yaml
├── .env.example
├── pyproject.toml
├── LAUNCH.md
└── README.mdLicense
MIT – Build freely, stay accountable.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
No tool schema history has been recorded yet.
This server cannot be installed
Maintenance
Related MCP Connectors
Deterministic runtime safety for AI agents: scan PII, gate tool actions, verify LLM output.
Zero-trust gateway for AI agents: score tool calls, verify agent cards, enforce policy, audit.
The WAF for agents. Pattern-based + heuristic firewall scans prompts, RAG documents, tool argume...
Deterministic validation for AI-generated artifacts: JSON Schema, OpenAPI response, SQL syntax.
Related MCP Servers
AlicenseAqualityDmaintenanceFact-checks and fixes AI outputs by catching hallucinations, repairing broken JSON, and correcting errors before they reach users, with tools for verification, validation, and correction.473MIT- AlicenseNot gradedqualityCmaintenanceProtects AI agents from prompt injection attacks, jailbreak attempts, and common web vulnerabilities by screening untrusted input through semantic LLM analysis and static pattern matching.482ISC
- AlicenseBqualityCmaintenanceValidates agent outputs in multi-agent systems to prevent coordination failures, with tools for schema verification, hallucination detection, and freshness checks, all with zero LLM cost.564MIT
- AlicenseNot gradedqualityCmaintenanceExposes LLM output validation (schema compliance and prompt-injection detection) as an MCP tool, enabling any MCP-compatible agent to apply real-time guardrails without importing the codebase.MIT
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- 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/princeruhulofficial/mcp-tool-guardian'
If you have feedback or need assistance with the MCP directory API, please join our Discord server