beast-agent@2.36.0
Ultra-fast local agent shell for Windows.
AI Security Review
scanned 1mo ago · by lpm-firewall-aiThe source contains risky automatic installation and broadly empowered agent functionality, but it does not meet the policy threshold for malicious intent or a publish block. Warn so the lifecycle downloader and remote content paths receive additional scrutiny.
Decision evidence
public snapshot- The postinstall hook automatically runs a repair script.
- That repair deletes Electron files and runs Electron's installer, with a third-party mirror fallback.
- The runtime agent can download and unpack a Python distribution before running user or model-supplied Python.
- The skills store fetches remote skill content and writes it into the application's active skills directory.
- Agent commands are broadly auto-approved except for deletion operations.
- There is no runtime dependency on beast-agent itself, so no nested lifecycle chain is present.
- The install hook is a transparent Electron repair path and does not modify AI-agent configuration or unrelated project files.
- No source evidence shows credential harvesting, secret exfiltration, stealth persistence, or a hidden remote payload.
Source & flagged code
18 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 unpkgFindings
Affected versions and remediation
This report applies to beast-agent@2.36.0.
See all versions of beast-agent for other recorded verdicts.
- Review the evidence and your use of beast-agent@2.36.0 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.