@drawio/mcp
Flagged as possibly malicious Reviewed: expected behaviour
Bawwab's static rules matched a pattern it treats as critical. Read the review notes: a reviewed false alarm is marked as such.
Why it was flagged
The static rules marked this “Flagged as possibly malicious” because of:
install-script.remote: Install script fetches remote content (critical, 1× · first at src/postinstall.js:28)
- Kind
- MCP server
- Source
- npm
- Version
1.13.0- Popularity
- 45,405 npm downloads/week (2026-11-20)
- Link
- https://www.npmjs.com/package/@drawio/mcp/v/1.13.0
- Scanned
- 2026-11-20T22:15:29+00:00
- Scan time
- 9.3 s
Review note: Not malicious, but one supply-chain weak spot: the postinstall downloads draw.io's routing script from the vendor's CDN (viewer.diagrams.net) into ~/.cache/drawio-mcp without a hash check, and the server runs it later, so a compromised CDN or cache would reach you.
Findings (2)
install-script: Runs code at install time med
package.json:15
OWASP LLM LLM03 Supply ChainOWASP Agentic ASI04 Agentic Supply Chain VulnerabilitiesOWASP Agentic ASI05 Unexpected Code Execution (RCE)MITRE ATLAS AML.T0010.001 Supply Chain Compromise: SoftwareMITRE ATLAS AML.T0011.001 User Execution: Malicious Package
node src/postinstall.js
What this means: Install hooks run automatically with your privileges before you can review anything.
The rule's general advice: Read the script; prefer packages with no install hooks. (see the review note for this case)
Expected for what this tool does
The npm postinstall runs src/postinstall.js, which downloads draw.io's libavoid edge-routing script from the vendor's own CDN (viewer.diagrams.net) into ~/.cache/drawio-mcp to save a later download. It does not run the code at install; the server loads it at runtime for 'libavoid' routing. The download is not pinned to a hash, but the source is the publisher's own CDN and the behaviour is documented in the code.
Reviewed by Claude (AI-assisted review), 2026-11-21
install-script.remote: Install script fetches remote content critical
src/postinstall.js:28
OWASP LLM LLM03 Supply ChainOWASP Agentic ASI04 Agentic Supply Chain VulnerabilitiesOWASP Agentic ASI05 Unexpected Code Execution (RCE)MITRE ATLAS AML.T0010.001 Supply Chain Compromise: SoftwareMITRE ATLAS AML.T0011.001 User Execution: Malicious PackageCWE CWE-494 Download of Code Without Integrity Check
// Let pending fetch handles drain: forcing an immediate exit can crash libuv
What this means: An install hook that downloads content can deliver a payload that differs from what was reviewed.
The rule's general advice: Do not install; vendor the download or pin it by hash. (see the review note for this case)
Expected for what this tool does
The matched line is a comment in the same postinstall cache-priming script described above: it fetches draw.io's routing script from viewer.diagrams.net into a local cache, checks it contains 'AvoidRouting', and never fails the install. It does not execute the download at install time.
Reviewed by Claude (AI-assisted review), 2026-11-21
1 low-severity notes (informational, do not affect the result)
files.binary-unscanned: Binary files not analysed low
1 binary file not analysed; use --deep to run in the sandbox
What this means: Prebuilt executables cannot be read by static rules, so what they do is unknown until they run.
What you can do: Use --deep to run the package in the sandbox, or prefer packages built from source.
Dispute this result
Maintainer and think this is wrong? Open a dispute on GitHub with the target id mcp-npm-drawio-mcp-75993a.