@sciagent/cli@1.1.31
SciAgent CLI - AI Research Assistant
AI Security Review
scanned 3mo ago · by lpm-firewall-aiThis is a concrete install-time remote-payload bootstrap chain with no artifact-integrity verification, but the inspected source does not establish data theft, stealth, or a foreign AI-agent control-surface mutation. It warrants a warning rather than a malicious block under the stated boundary.
Decision evidence
public snapshot- package.json runs scripts/postinstall.js automatically.
- postinstall downloads unsigned platform binaries into ~/.sciagent/bin.
- postinstall fetches latest CodeBuddy wheel metadata and extracts its binary.
- Fallback runs `npm install -g @tencent-ai/codebuddy-code` during install.
- bin/sciagent.js later executes the downloaded sciagent binary and auto-updates it.
- No credential harvesting or outbound local-data upload found.
- No AI-agent configuration/control-surface writes found.
- Shell commands are limited to download, extraction, process, and binary-management tasks.
Source & flagged code
6 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 @sciagent/cli@1.1.31.
See all versions of @sciagent/cli for other recorded verdicts.
- Review the evidence and your use of @sciagent/cli@1.1.31 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.