registry / Data exfiltration / @hashbrownai/openai
@hashbrownai/openai
OpenAI provider for Hashbrown AI
This report shows the evidence for @hashbrownai/openai@0.8.2. 1 of 8 known versions is blocked or flagged. The newest version, 0.8.3, is flagged. See all versions.
AI Security Review
scanned 6d ago · by lpm-firewall-aiThe default postinstall sends Git display names and email addresses to the vendor endpoint, which is install-time disclosure of user data without consent. The runtime entry point is separate, but that does not change the automatic install behavior.
Static reason
One or more suspicious static signals were detected.; source matched previously finalized malicious package; routed for review
Trigger
Package installation runs the postinstall hook by default.
Impact
User identity and project context are disclosed to the package vendor during installation.
Mechanism
The hook reads Git configuration, adds the user's display name and email to the install event, and posts it to hashbrown.dev unless disabled by an environment variable.
Attack narrative
During installation, the package automatically runs a collector that reads Git configuration and includes the user's Git display name and email in an install event. It sends that event to hashbrown.dev. An opt-out variable exists, but the collection is enabled by default and includes personal and project-related data beyond anonymous install telemetry.
AI rationale
The default postinstall sends Git display names and email addresses to the vendor endpoint, which is install-time disclosure of user data without consent. The runtime entry point is separate, but that does not change the automatic install behavior.
Evidence
package.jsoninstall/postinstall.cjsinstall/collector.cjsinstall/git-context.cjsinstall/policy.cjsinstall/identity.cjs~/.hashbrown/installation-id{packageRoot}/development-install.mjs{packageRoot}/development-install.cjs
Network endpoints1
hashbrown.dev/_/growth/collect/v1/install
Decision evidence
public snapshotAI called this Malicious at 97.0% confidence as Malware with low false-positive risk.
Evidence for block
- The manifest runs install/postinstall.cjs automatically during installation.
- The collector reads Git configuration and collects the user's Git display name and email.
- The install event includes collected identity outside CI and sends the batch to the configured endpoint.
- The default collection endpoint is https://hashbrown.dev/_/growth/collect/v1/install.
Evidence against
- The package's runtime entry point only exports the OpenAI adapter; the collector is not imported there.
- The collector supports telemetry opt-out environment variables.
- The installation identifier is randomly generated rather than read from an existing service credential.
Behavioral surface
CryptoEnvironmentVarsFilesystemNetwork
UrlStrings
Source & flagged code
7 flaggedpackage.jsonView file
•scripts.postinstall = node install/postinstall.cjs
High
Install Time Lifecycle Scripts
Package defines install-time lifecycle scripts.
package.jsonView on unpkg36"scripts": {
"postinstall": "node install/postinstall.cjs"
}
}
High
Suspicious Lifecycle Evidence
The manifest runs install/postinstall.cjs automatically during installation.
package.jsonView on unpkg · L36•scripts.postinstall = node install/postinstall.cjs
Medium
Ambiguous Install Lifecycle Script
Install-time lifecycle script is not statically allowlisted and needs review.
package.jsonView on unpkginstall/git-context.cjsView file
Findings
6 High3 Medium3 Low
HighInstall Time Lifecycle Scriptspackage.json
HighKnown Malware Source Similarityinstall/git-context.cjs
HighKnown Malware Source Similarityinstall/policy.cjs
HighSuspicious Lifecycle Evidencepackage.json
HighSuspicious Lifecycle Evidenceinstall/postinstall.cjs
HighAi Review Evidenceinstall/git-context.cjs
MediumAmbiguous Install Lifecycle Scriptpackage.json
MediumNetwork
MediumEnvironment Vars
LowScripts Present
LowFilesystem
LowUrl Strings
Affected versions and remediation
This report applies to @hashbrownai/openai@0.8.2.
The versions list below shows the verdict for each recorded version.
- Avoid installing @hashbrownai/openai@0.8.2. 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: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.