hippo-memory@1.52.6
Memory for AI agents that learns what is wrong and stops repeating it. MCP server and hooks for Claude Code, Codex and Cursor. SQLite, zero runtime deps.
AI Security Review
scanned 14d ago · by lpm-firewall-aiThe automatic postinstall behavior includes a guarded repair of an existing, previously opted-in Codex wrapper, which is a limited launcher lifecycle risk. Agent hook setup is separately reached through explicit CLI initialization; the inspected source does not establish a concrete malicious chain.
Decision evidence
public snapshot- The package runs a postinstall script, which loads the compiled postinstall module.
- Postinstall attempts to repair the Codex launcher wrapper only when existing metadata records prior user opt-in.
- The wrapper integration targets the user's Codex launcher path; this is a guarded same-command wrapper repair.
- Agent hook installation is invoked by the explicit `hippo init` command, not by postinstall.
- Postinstall prints setup guidance and describes Claude detection as read-only.
- No source evidence inspected establishes secret collection, unrelated data exfiltration, destructive behavior, or remote payload execution.
Source & flagged code
7 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/embeddings.jsView on unpkg · L99Findings
Affected versions and remediation
This report applies to hippo-memory@1.52.6.
See all versions of hippo-memory for other recorded verdicts.
- Review the evidence and your use of hippo-memory@1.52.6 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.