@sciagent/cli@1.1.85
SciAgent CLI - AI Research Assistant
AI Security Review
scanned 3mo ago · by lpm-firewall-aiThe package is a remote binary bootstrapper with install-time global dependency installation and no artifact verification. Its risky behavior is package-aligned, but it remains an unresolved staged-payload capability rather than confirmed malware.
Decision evidence
public snapshot- package.json runs postinstall.
- scripts/postinstall.js downloads unsigned platform binaries during install.
- postinstall globally installs @tencent-ai/codebuddy-code as fallback.
- bin/sciagent.js executes downloaded binary and performs background updates.
- Downloads are persisted under ~/.sciagent/bin or LOCALAPPDATA/sciagent/bin.
- No source reads user credentials, project files, or AI-agent configuration.
- No observed exfiltration, destructive action, eval, or foreign agent-control writes.
- Network paths are dedicated to declared SciAgent/CodeBuddy binary delivery.
Source & flagged code
7 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 unpkgSource gates dangerous network, credential, or execution behavior behind CI, host, platform, time, or geo fingerprint checks.
scripts/postinstall.jsView on unpkg · L8Install-named source file stages remote content through filesystem writes and execution.
scripts/postinstall.jsView on unpkg · L8Package source invokes a package manager install command at runtime.
scripts/postinstall.jsView on unpkg · L485Findings
Affected versions and remediation
This report applies to @sciagent/cli@1.1.85.
See all versions of @sciagent/cli for other recorded verdicts.
- Review the evidence and your use of @sciagent/cli@1.1.85 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.