@jait/gateway@0.1.871
Jait AI gateway — local-first AI coding agent with terminal, filesystem, and browser control
AI Security Review
scanned 14d ago · by lpm-firewall-aiThe source supports a first party agent integration capability that is activated by starting a Pi session, so this warrants a warning under the agent integration policy. The reported browser credential behavior is a configured API client request, not evidence of theft.
Decision evidence
public snapshot- Starting a Pi based ACP session conditionally provisions a per session MCP configuration and wrapper.
- The provisioner writes the supplied MCP server configuration and a Pi wrapper in a temporary directory.
- The source describes this configuration as keeping session scoped URLs out of global and project Pi configuration.
- The browser search tool sends its configured OpenAI key to the configured OpenAI base URL, defaulting to the OpenAI API; the request is validated by an SSRF guard, so the scanner hint alone does not establish credential theft.
- The inspected agent setup is activated by starting a Pi based session and uses temporary session scoped files; no persistent agent configuration write was found in these excerpts.
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@0.1.871.
See all versions of @jait/gateway for other recorded verdicts.
- Review the evidence and your use of @jait/gateway@0.1.871 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.