thalamus@1.5.106
Mobile and Web client for Claude Code
AI Security Review
scanned 4d ago · by lpm-firewall-aiThe reviewed flagged behavior has an ordinary purpose: postinstall extracts bundled tools, OAuth usage data goes to Anthropic, and GitHub credentials are scoped to GitHub. Vendor token registration targets the configured Thalamus service; no unrelated exfiltration or concealed payload execution was confirmed.
Decision evidence
public snapshot- The postinstall script unpacks bundled platform tools from package archives and does not contact a network host.
- Anthropic OAuth usage requests send the token to Anthropic's own API endpoint.
- The Git credential helper is scoped to github.com; vendor token registration uses the configured Thalamus server URL.
- The package has a postinstall hook that writes unpacked tools into tools/unpacked.
- The source includes process execution for launching the CLI and helper tools.
Source & flagged code
20 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@1.5.106.
See all versions of thalamus for other recorded verdicts.
- Review the evidence and your use of thalamus@1.5.106 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.