trace-mcp@3.23.0
Framework-aware code intelligence MCP server — 87 framework integrations, 81 languages, 70.5% fewer input tokens to review a pull request, comprehension at parity
AI Security Review
scanned 2mo ago · by lpm-firewall-aiThe package has a concrete automatic persistence and runtime self-dependency chain, but the inspected code is readable and package-aligned rather than covertly exfiltrating data or modifying a foreign AI-agent configuration. It warrants a warning rather than a publication block.
Decision evidence
public snapshot- The install lifecycle automatically runs three package scripts.
- The postinstall handler writes and bootstraps a per-user macOS LaunchAgent.
- Normal server use schedules an automatic self-update that runs npm to install the latest trace-mcp release.
- The inspected lifecycle scripts are readable source, not hidden or encoded payloads.
- Claude hook configuration is exposed through an explicit setup-hooks command rather than the install hook.
- The self-update checks a fixed npm registry endpoint and has an environment opt-out.
Source & flagged code
17 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgInstall-time lifecycle script is not statically allowlisted and needs review.
package.jsonView on unpkgFindings
Affected versions and remediation
This report applies to trace-mcp@3.23.0.
See all versions of trace-mcp for other recorded verdicts.
- Review the evidence and your use of trace-mcp@3.23.0 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.