@jive-sh/skill-sync@0.1.1
⚠ Under reviewKeeps one skills directory identical across every machine you work on.
AI Security Review
scanned 14d ago · by lpm-firewall-aiThe inspected source confirms install-time daemon registration, while the platform-specific executable selected by that installer could not be reviewed. Because this unresolved registry dependency is required to assess the daemon’s behavior, the package needs manual review; the available source does not establish malicious activity.
Decision evidence
public snapshot- The postinstall hook runs install.js, which registers the package daemon to start at login unless CI or repair-only mode skips registration.
- The package declares platform-specific aliases of itself as optional dependencies, and the installer uses the platform binary as the daemon executable.
- The installer records the executable and registration details; the installed daemon’s behavior could not be verified because the selected platform package source was unavailable.
- The inspected config code selects a local skills directory, but this does not establish what the unavailable daemon does with it.
- The source inspected here does not establish credential theft, data exfiltration, destructive behavior, or an attack against agent configuration.
- Commander, the inspected registry dependency, has no install scripts in its prepared manifest.
Source & flagged code
6 flaggedPackage defines install-time lifecycle scripts.
package.jsonView on unpkgInstall-time lifecycle script is not statically allowlisted and needs review.
package.jsonView on unpkgFindings
Affected versions and remediation
This report applies to @jive-sh/skill-sync@0.1.1.
See all versions of @jive-sh/skill-sync for other recorded verdicts.
- Review the evidence and your use of @jive-sh/skill-sync@0.1.1 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.