thalamus@1.5.46
⚠ Under reviewMobile and Web client for Claude Code
AI Security Review
scanned 29d ago · by lpm-firewall-aiThe inspected source shows a local archive extraction hook and benign Claude transcript path handling, but the flagged binary archive could not be examined. Manual review is appropriate to resolve its contents; the available evidence does not establish malicious behavior.
Decision evidence
public snapshot- The postinstall hook runs a local script that extracts bundled platform tools into the package’s tools directory.
- The agent-configuration hint resolves to code that constructs a Claude project transcript path and reads that transcript.
- The bundled archive is binary and could not be inspected through the available source reader, leaving its contents unresolved.
- The inspected postinstall code extracts a named platform archive into tools/unpacked; it shows no network request or execution of the extracted files.
- The MCP launcher starts the package’s own bridge entrypoint.
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@1.5.46.
See all versions of thalamus for other recorded verdicts.
- Review the evidence and your use of thalamus@1.5.46 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.