@mindflight/ghostcrab-personal-mcp@0.7.1
SQLite-backed MCP memory server (MindBrain + Zig backend), part of the MindBrain ecosystem.
AI Security Review
scanned 1mo ago · by lpm-firewall-aiThe install hook performs non-consensual consumer-project and existing agent-config mutation. Its scope appears limited to this product and no exfiltration or self-dependency chain was found, so a warning is more proportionate than a block.
Decision evidence
public snapshot- The manifest runs a postinstall script automatically.
- Postinstall creates consumer .env and data paths and adds package-controlled documentation and skill symlinks.
- Postinstall invokes an upgrade routine without dry-run or process-kill opt-out.
- That routine rewrites existing Cursor and Codex MCP configuration to register this package's server.
- The configuration entries are specifically named for GhostCrab rather than arbitrary servers.
- No runtime dependency on this package's own name was found.
- No credential harvesting or external-network action was found in the install path.
Source & flagged code
16 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgPackage source references child process execution.
bin/lib/prebuild-permissions.mjsView on unpkg · L11Findings
Affected versions and remediation
This report applies to @mindflight/ghostcrab-personal-mcp@0.7.1.
See all versions of @mindflight/ghostcrab-personal-mcp for other recorded verdicts.
- Review the evidence and your use of @mindflight/ghostcrab-personal-mcp@0.7.1 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.