@hashbrownai/ollama@0.8.2
Ollama provider for Hashbrown AI
AI Security Review
scanned 6d ago · by lpm-firewall-aiThe default postinstall hook sends Git identity and checkout context to the vendor’s endpoint without explicit consent. Since the event includes personal and project details from the user’s machine, this is data exfiltration.
Decision evidence
public snapshot- The package runs install/postinstall.cjs automatically from its postinstall lifecycle script.
- The postinstall handler calls collectInstall during npm postinstall.
- The default endpoint is https://hashbrown.dev/_/growth/collect/v1/install; only loopback endpoint overrides are accepted.
- The collector adds identity data to the install event and sends the batch to the selected endpoint.
- Git discovery reads configured Git name and email and includes them as gitDisplayName and gitEmail in identity.
- The event also contains ordinary package installation telemetry such as package version, OS family, architecture, Node version, and package-manager details.
- The collector supports telemetry opt-out environment variables.
Source & flagged code
7 flaggedPackage defines install-time lifecycle scripts.
package.jsonView on unpkgThe package runs install/postinstall.cjs automatically from its postinstall lifecycle script.
package.jsonView on unpkg · L36Install-time lifecycle script is not statically allowlisted and needs review.
package.jsonView on unpkgFindings
Affected versions and remediation
This report applies to @hashbrownai/ollama@0.8.2.
See all versions of @hashbrownai/ollama for other recorded verdicts.
- Avoid installing @hashbrownai/ollama@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.