trace-mcp@3.25.0
Framework-aware code intelligence MCP server — 88 framework integrations, 81 languages, 72.7% fewer input tokens to review a pull request, comprehension at parity
AI Security Review
scanned 1mo ago · by lpm-firewall-aiThe package has a real, automatic persistence and remote-update surface, but it is scoped to trace-mcp and does not meet the policy threshold for a malicious publish block. Treat it as a lifecycle risk requiring warning.
Decision evidence
public snapshot- The automatic postinstall chain runs two system-mutating scripts.
- On macOS it writes and bootstraps a per-user launchd agent configured to run at load and stay alive.
- Postinstall downloads a GitHub release, verifies only a release-provided checksum, and can replace installed trace-mcp app bundles.
- The shipped Codex hook runs a package shell guard before Bash tool use when the plugin is enabled.
- No runtime dependency on trace-mcp itself is declared.
- The persistence target and app paths are scoped to trace-mcp rather than a broad agent configuration.
- The screenshot and tool-count scripts are explicit developer commands, not lifecycle hooks.
- No credential harvesting or outbound secret transmission was found in the inspected lifecycle sources.
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@3.25.0.
See all versions of trace-mcp for other recorded verdicts.
- Review the evidence and your use of trace-mcp@3.25.0 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.