@jait/gateway@0.1.865
⚠ Under reviewJait AI gateway — local-first AI coding agent with terminal, filesystem, and browser control
AI Security Review
scanned 14d ago · by lpm-firewall-aiThe inspected source does not substantiate the scanner’s credential-exfiltration claim or a concrete attack. Required clean coverage exceeds the citation limit and includes a binary path, so the package cannot be marked clean under the supplied review rules.
Decision evidence
public snapshot- The package has a postinstall hook that runs a runtime provisioning script.
- Web-search keys come from configured keys, environment variables, or matching Jait backend instances.
- The OpenAI search path validates its selected endpoint before sending the key in an authorization header.
- Pi MCP provisioning uses a per-session temporary configuration and wrapper.
- The inspected source does not establish credential theft or an unconsented agent configuration write.
- The required clean review includes 20 paths, exceeding the citation limit; one required path is binary and cannot be cited as UTF-8 source.
Source & flagged code
27 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgFindings
Affected versions and remediation
This report applies to @jait/gateway@0.1.865.
See all versions of @jait/gateway for other recorded verdicts.
- Review the evidence and your use of @jait/gateway@0.1.865 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.