thalamus@1.5.79
⚠ Under reviewMobile and Web client for Claude Code
AI Security Review
scanned 20d ago · by lpm-firewall-aiInspected source shows ordinary path helpers, package-aligned API calls, and explicitly gated loopback-only log shipping. The postinstall-extracted binary archive remains unresolved, so the available evidence does not support either a clean verdict or a confirmed attack.
Decision evidence
public snapshot- package.json runs scripts/unpack-tools.cjs after install; the script selects and extracts a platform archive into tools/unpacked.
- All four flagged .claude locations construct project paths or process extracted text; they do not show agent configuration mutation.
- The API uses the Thalamus server by default and sends session metadata encrypted in its session request.
- Remote log shipping requires an explicit environment opt-in and is restricted to a loopback server.
- The shipped difftastic archive is binary, and its contents could not be resolved through source inspection.
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@1.5.79.
See all versions of thalamus for other recorded verdicts.
- Review the evidence and your use of thalamus@1.5.79 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.