codesift-mcp@0.23.0
MCP server for code intelligence — 146 tools for symbol search, call graph, semantic search, route tracing, community detection, LSP bridge, secret detection, conversation search, and Hono framework intelligence
AI Security Review
scanned 12d ago · by lpm-firewall-aiThe postinstall hook automatically invokes setup across platforms, which can modify agent configuration and rules. The inspected behavior is package-aligned first-party setup, so it warrants a warning rather than a malicious verdict.
Decision evidence
public snapshot- The postinstall hook automatically runs setup for all supported platforms.
- Codex setup writes configuration under the user’s .codex directory and ensures an approval setting for the package’s own server entry.
- The rules setup maps platforms to package-owned files and appends managed rules to the project’s AGENTS.md or GEMINI.md.
- The package configuration behavior is an agent setup capability; no separate concrete attack was established from the inspected source.
- The inspected setup code targets CodeSift’s own MCP entry and uses marked package-owned rules blocks.
- No network endpoint or unrelated dependency approval grant was established in the inspected excerpts.
Source & flagged code
23 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 unpkgPackage source references a known benign dynamic code generation pattern.
dist/storage/sqlite/native-sqlite.jsView on unpkg · L70Findings
Affected versions and remediation
This report applies to codesift-mcp@0.23.0.
See all versions of codesift-mcp for other recorded verdicts.
- Review the evidence and your use of codesift-mcp@0.23.0 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.