thalamus@1.5.52
⚠ Under reviewMobile and Web client for Claude Code
AI Security Review
scanned 18d ago · by lpm-firewall-aiStatic inspection did not verify malicious behavior in the flagged agent-related code or the managed MCP installer. The postinstall extracts an executable archive whose contents could not be verified, leaving a material question unresolved; manual review is appropriate.
Decision evidence
public snapshot- package.json runs scripts/unpack-tools.cjs during postinstall.
- The installer extracts the platform-specific difftastic archive and makes extracted files executable; the archive contents could not be verified in this inspection.
- The flagged agent-related locations read Claude settings paths or derive a project transcript path; these excerpts do not establish agent-control mutation.
- The managed MCP installer disables dependency lifecycle scripts when it invokes npm.
- The archive is selected by platform and unpacked as a tool asset; the inspected caller does not execute its contents during extraction.
- MCP server launch configuration is loaded from the organization registry, and only enabled entries are used.
Source & flagged code
16 flaggedPackage defines install-time lifecycle scripts.
package.jsonView on unpkgPackage source references child process execution.
bin/thalamus-mcp.mjsView on unpkg · L2Findings
Affected versions and remediation
This report applies to thalamus@1.5.52.
See all versions of thalamus for other recorded verdicts.
- Review the evidence and your use of thalamus@1.5.52 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.