@phnx-labs/agents-cli@1.22.101
One CLI for all your AI coding agents - versions, config, cloud dispatch, sessions, and teams (now with first-class Grok Build CLI support)
AI Security Review
scanned 1mo ago · by lpm-firewall-aiThe package has an automatic lifecycle hook that creates user-level persistence and modifies command lookup. The behavior appears first-party and package-aligned rather than a concrete malicious payload chain.
Decision evidence
public snapshot- The npm postinstall hook runs automatically after installation.
- On POSIX, postinstall creates agents and ag symlinks and can append ~/.local/bin to a login shell profile without an interactive prompt.
- Postinstall starts or restarts an always-on daemon on macOS and Linux.
- The daemon registers and starts a per-user launchd or systemd service.
- The manifest has no runtime dependency on this package itself.
- The installer limits its direct setup to this CLI's paths and shims; it does not modify third-party agent configuration.
- Its internal self-update installation explicitly uses --ignore-scripts.
- No install-time network request or credential exfiltration was found in the inspected lifecycle path.
Source & flagged code
22 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgInstall-time lifecycle script is not statically allowlisted and needs review.
package.jsonView on unpkgPackage source references dynamic require/import behavior.
dist/lib/sqlite.jsView on unpkg · L15Findings
Affected versions and remediation
This report applies to @phnx-labs/agents-cli@1.22.101.
See all versions of @phnx-labs/agents-cli for other recorded verdicts.
- Review the evidence and your use of @phnx-labs/agents-cli@1.22.101 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.