thalamus@1.5.74
Mobile and Web client for Claude Code
AI Security Review
scanned 21d ago · by lpm-firewall-aiThe source supports a warning for first-party agent extension setup: eligible runtime sessions load bundled workflow skills through a plugin directory. The inspected postinstall hook unpacks platform tools, and the supplied integration locations do not establish a separate attack.
Decision evidence
public snapshot- The postinstall hook runs a tool unpacker that extracts platform-specific archives.
- The four supplied agent-integration locations concern Claude project paths and friction-metrics state.
- Runtime code writes bundled workflow skills as a plugin under Thalamus's home directory.
- Eligible session types pass that plugin directory to Claude when launching a session.
- The unpacking code selects archives by the current platform.
Source & flagged code
18 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@1.5.74.
See all versions of thalamus for other recorded verdicts.
- Review the evidence and your use of thalamus@1.5.74 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.