Atlas / MCP servers / jujumilk3 / Ezmcp

EzmcpSAFE

mcp/jujumilk3/ezmcp

Easy-to-use MCP server framework specialized for SSE.

Verdict
SAFE
Grade
B
Trust score
89 /100
Exposed tools
12 10r · 1w · 1d
Transport
—
License
—
Stars
23
01

Overview

From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.

Easy-to-use MCP server framework specialized for SSE.

Overview

ezmcp is a lightweight framework that makes it easy to create MCP-compatible servers using a FastAPI-like syntax. It provides a simple decorator-based API for defining tools that can be called by MCP clients.

Features

  • FastAPI-style decorator API for defining MCP tools
  • Automatic parameter validation and type conversion
  • Automatic generation of tool schemas from function signatures
  • Built-in support for SSE transport
  • FastAPI-style middleware support
  • Easy integration with existing Starlette applications
  • Interactive documentation page for exploring and testing tools

Installation

pip install ezmcp

Quick Start

from ezmcp import ezmcp, TextContent

# Create an ezmcp application
app = ezmcp("my-app")

# Define a tool
@app.tool(description="Echo a message back to the user")
async def echo(message: str):
"""Echo a message back to the user."""
return [TextContent(type="text", text=f"Echo: {message}")]

# Run the application
if __name__ == "__main__":
app.run(host="0.0.0.0", port=8000)

Once the server is running, you can:

  • Access the interactive documentation at http://localhost:8000/docs
  • Connect to the SSE endpoint at http://localhost:8000/sse

Middleware

ezmcp supports middleware similar to FastAPI, allowing you to add behavior that is applied across your entire application.

from starlette.requests import Request

from ezmcp import TextContent, ezmcp

app = ezmcp("my-app")

@app.middleware
async def process_time_middleware(request: Request, call_next):
"""Add a header with the processing time."""
import time
start_time = time.perf_counter()
response = await call_next(request)
process_time = time.perf_counter() - start_time
response.headers["X-Process-Time"] = str(process_time)
return response

@app.tool(description="Echo a message back to the user"
Read from source at commit d64a256cb9f6OBSERVED · 2026-10-09
02

Connect

Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.

claude-code
claude mcp add ezmcp -- uvx ezmcp
claude-desktop
{
  "mcpServers": {
    "ezmcp": {
      "command": "uvx",
      "args": [
        "ezmcp"
      ]
    }
  }
}
03

Exposed tools (12)

10 read · 1 write · 1 destructive. Blast radius: 1 tool can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.

ToolRiskDescription
addwriteAdd two numbers together.
delete_memorydestructive
echoreadEcho a message back to the user.
fetch_imagereadFetch a sample image.
fetch_websitereadFetch a website and return its content.
greetreadGet a greeting with the user
list_memoriesread
retrieve_memoryread
search_memoriesread
store_memoryread
test_toolreadreturn [TextContent(type=
user_inforeadReturn user information as JSON.
04

Trust audit

SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.

LayerWhat it checksResult
L0Provenance & inventoryPASS
L1Static analysis of the codePASS
L2Instruction surface (what it tells the agent)PASS
L3Class-specific surfaceWARN
L4Behavioural (sandbox)SKIPPED

What the source does

Filesystem
declared (1 observation(s))
Network
none-observed
Shell
none-observed
Dependencies
pinned
Secrets in source
none-found

Findings (2)

MEDIUMFilesystem / path · mcp.destructive_tools · CWE-22, CWE-59
delete_memory
Why it matters. 1 tool(s) can delete or overwrite
Fix. prefer a read-only mode or scoped tokens; the page states the blast radius
LOWInventory / provenance · inv.no_license · CWE-1104
Why it matters. no LICENSE file and no repo licence
Fix. add a licence

Gates applied: no_behavioural_pass, no_license.

Audited 2026-10-09 · audit v0.4.1 · source sha d64a256cb9f6full audit observations/trust-audit/mcp-server/jujumilk3__ezmcp.json · Report an issue / request a re-scan
05

Audit history

Every audit this server has had. A grade with a past is a grade somebody is still checking.

DateSourceVerdictGradeScoreChange
2026-10-09d64a256cb9f6SAFEB89first audit
06

Questions

What is the Ezmcp MCP server?

Easy-to-use MCP server framework specialized for SSE.

What tools does Ezmcp expose?

12 in total: 10 read-only, 1 that write, and 1 that can delete or overwrite (delete_memory). Every one is listed on this page with its risk.

Is Ezmcp safe to connect to an agent?

The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B. Separately from the audit: 1 of its tools can destroy data, so scope the token you give it to what you actually need.

What credentials does Ezmcp need?

No credential environment variables were found in its source, so it appears to need none.

How current is this page?

The grade is for one exact copy of the source (d64a256cb9f6), read on 2026-10-09. The repository is watched and re-audited when it changes.

Advertisement