@solongate/proxy@0.90.56
AI tool security proxy: protect any AI tool server with customizable policies, path/command constraints, rate limiting, and audit logging. No code changes required.
AI Security Review
scanned 2mo ago · by lpm-firewall-aiThe package has no automatic npm lifecycle hook, but its explicit global setup installs broad agent hooks that export conversation content and accept remotely supplied replacement code. This is a concrete persistent data-exfiltration and remote-code-update chain. Product guard normalized a non-low false-positive publish_block request to warn-only suspicious.
Decision evidence
public snapshot- Its global installer registers command hooks for every Codex tool event, including user prompts and session stop events.
- The conversation hook sends redacted user prompt and assistant-message content to api.solongate.com using the configured API key.
- The installed guard periodically downloads JavaScript hook content from the service and atomically replaces local hook files.
- The installer writes global hook configuration and copies hook programs into the user home directory.
- package.json has no preinstall, install, or postinstall lifecycle hook.
- The network destination is the package vendor's API and the behavior is reachable through its CLI setup path.
Source & flagged code
27 flagged · loading sourcePackage source references child process execution.
dist/tui/index.jsView on unpkg · L143Findings
Affected versions and remediation
This report applies to @solongate/proxy@0.90.56.
See all versions of @solongate/proxy for other recorded verdicts.
- Review the evidence and your use of @solongate/proxy@0.90.56 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.