@mindflight/ghostcrab-personal-mcp
SQLite-backed MCP memory server (MindBrain + Zig backend), part of the MindBrain ecosystem.
This report shows the evidence for @mindflight/ghostcrab-personal-mcp@0.6.9. 4 of 9 known versions are blocked or flagged. The newest version, 0.7.4, is flagged. See all versions.
AI Security Review
scanned 1mo ago · by lpm-firewall-aiThis is malicious under the install-control-surface policy because a postinstall hook automatically rewrites existing global Cursor and Codex agent configuration. The additional consumer-project mutations confirm broad automatic install-time modification.
Decision evidence
public snapshot- The manifest runs a postinstall script automatically.
- Postinstall automatically invokes an upgrade routine with configuration cleanup enabled.
- That routine rewrites existing global Cursor and Codex MCP configuration to launch this package.
- Postinstall also creates consumer-project files and links without an explicit command.
- No runtime dependency on this package's own name was found.
- The reviewed configuration writes point to the package's own MCP server and no secret exfiltration endpoint was identified.
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@0.6.9.
The versions list below shows the verdict for each recorded version.
- Avoid installing @mindflight/ghostcrab-personal-mcp@0.6.9. Remove it from direct dependencies and check your lockfile for transitive copies.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
- If this version ran, investigate the affected machine and build environment. Rotate credentials it could access and rebuild from a trusted environment.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.