Secure MCP Server Template
Allows executing read-only SQL queries against a PostgreSQL database, returning rows as JSON.
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
Secure MCP Server Template
A minimal, correct, and security-first Model Context Protocol (MCP) server in Python. It exposes three example tools to any MCP client (Claude Desktop, Claude Code, etc.) — each one shipped with the security guard it actually needs.
Most "MCP server" tutorials show you how to expose a tool. They skip the part that matters once a model has real hands on your database, network, and disk. This template is built around that gap: how to give an LLM capabilities without opening a hole.
LLM client ──MCP──> this server ──> [ PostgreSQL | HTTP | filesystem ]
│
└── every tool passes through guards.py (tested)What is MCP, in one paragraph
MCP is an open protocol that lets an AI client call external "tools" (functions) and read "resources" through a standard interface. You run a small server that declares tools; the client (Claude Desktop, Claude Code, …) discovers them and the model calls them during a conversation. It's the clean way to give a model access to your data and actions instead of pasting everything into the prompt.
Related MCP server: KingBuilds MCP Server
Tools
Tool | What it does | Guard |
| Run a read-only SQL query against | SELECT/WITH only and a real |
| HTTP GET, returns the body as text | http/https only + host allowlist, deny-by-default (anti-SSRF) |
| Read a text file under | Path resolved inside the root, blocks |
Security design (the point of this template)
db_querycan't write. Two layers: a static check (is_safe_select— rejects writes, DDL, and stacked;statements) and the connection is openedREAD ONLY, so even a clever bypass can't mutate data.http_getcan't be turned into an SSRF. Onlyhttp(s), and the host must be inMCP_URL_ALLOWLIST. Empty allowlist = nothing allowed (deny-by-default), so a misconfigured server isn't an open proxy.read_filecan't escape its root. Paths are resolved and checked to live insideMCP_FILES_ROOT;../, absolute paths, and symlink escapes raise.
All three guards live in guards.py (zero dependencies) and are covered by tests/test_guards.py, so they run without even installing MCP.
Quickstart
git clone <this repo> && cd mcp-server-template
python -m venv .venv && . .venv/Scripts/activate # (Linux/mac: . .venv/bin/activate)
pip install -e . # add: pip install -e ".[db]" for PostgreSQL
cp .env.example .env # edit with your values
python server.py # runs over stdioUse it from Claude Desktop
Add to your claude_desktop_config.json (see claude_desktop_config.example.json):
{
"mcpServers": {
"atelier-template": {
"command": "python",
"args": ["/absolute/path/to/server.py"],
"env": { "MCP_URL_ALLOWLIST": "api.github.com", "MCP_FILES_ROOT": "/safe/dir" }
}
}
}Restart Claude Desktop; the three tools appear. (For Claude Code: claude mcp add atelier-template -- python /abs/path/server.py.)
Add your own tool (the 5-line version)
@mcp.tool()
def word_count(text: str) -> str:
"""Count words in a string.""" # <- this docstring is what the model reads
return str(len(text.split()))That's the whole loop: decorate a function, write a clear docstring (the model uses it to decide when/how to call), return a string. If the tool touches data/network/disk, add a guard — that's the habit this template is trying to teach.
Layout
server.py # the MCP server + 3 example tools
guards.py # pure, dependency-free security helpers (tested)
tests/ # pytest for the guards (run without mcp installed)
.env.example # configurationTests
pip install pytest && pytest -qLicense
MIT © 2026 Eduardo Pérez Ignacio
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
This server cannot be deployed
Maintenance
Related MCP Connectors
Safe, read-only Postgres and MySQL access for AI agents. Audit log + column-level controls.
Read authorized Loam tables and supported external databases through a bounded MCP connection.
Read-only cited financial datasets, bounded research queries and source diagnostics.
Free bounded public HTTPS document and JSON reads. Exact host allowlist; real source provenance.
Related MCP Servers
- AlicenseNot gradedqualityBmaintenanceProvides secure read-only SQL access to PostgreSQL and ClickHouse databases with built-in safety features like read-only enforcement, timeouts, and managed result files.63 PyPI1MIT
- AlicenseAqualityDmaintenanceProvides sandboxed file operations, web scraping, data transformation, and authenticated HTTP requests with SSRF protection.6MIT
- AlicenseNot gradedqualityDmaintenanceProvides authenticated, rate-limited file reading, directory listing, and HTTPS fetching with strict path and domain allowlists.MIT
- FlicenseNot gradedqualityCmaintenanceEnables read-only access to explicitly approved local and remote filesystems, including inspecting Git repositories and querying SQLite and DuckDB databases, with root administration kept separate from MCP.-