superlocalmemory@4.2.6
The governed memory brain for AI agents: one local memory for Claude Code, Codex, Cursor, Copilot, ChatGPT and any MCP client. Honest recall, documents and pictures, roles, audit and GDPR erasure.
AI Security Review
scanned 8d ago · by lpm-firewall-aiSource inspection found ordinary package runtime setup in postinstall and user-invoked memory commands in the flagged hooks. The optional media request requires an affirmative choice, and no concrete attack behavior was established.
Decision evidence
public snapshot- package.json binds postinstall to scripts/postinstall.js.
- The installer creates a package-owned virtual environment and installs the matching superlocalmemory Python package into it.
- The list, recall, and remember hooks pass command-line arguments to their corresponding memory commands.
- The profile hook passes an explicit profile command and prompts before deletion or switching.
- The reset hook asks for confirmation before a soft reset.
- Optional media setup asks an interactive default-no question and records nothing when declined.
Source & flagged code
10 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 unpkgSource file is highly similar to a previously finalized malicious package; route for source-aware review.
Findings
Affected versions and remediation
This report applies to superlocalmemory@4.2.6.
See all versions of superlocalmemory for other recorded verdicts.
- Review the evidence and your use of superlocalmemory@4.2.6 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.