@nalvietnam/avatar-cli@3.24.1
AI harness CLI for NAL Vietnam engineering
AI Security Review
scanned 2mo ago · by lpm-firewall-aiThe package presents a real first-party AI-extension lifecycle risk through remotely supplied, softly verified packs. Source inspection found no concrete malware behavior, covert exfiltration, or unconsented install-time control-surface mutation.
Decision evidence
public snapshot- Explicit avatar init/sync downloads a remote team pack into .claude/pack/; the documented digest check is soft and permits installation without a digest.
- Runtime commands merge pack-provided settings/hooks and manage Claude workspace files, creating an agent-extension supply-chain capability.
- The install-time postinstall only imports and prints a welcome screen; no install-time agent-config mutation was found.
- No credential harvesting or unrelated command-output exfiltration was established.
- Remote endpoints are package-aligned Supabase auth/pack services, not covert third-party collectors.
- AI-control-surface writes are user-invoked CLI behavior, not preinstall/install/postinstall behavior.
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 unpkgThis package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
Findings
Affected versions and remediation
This report applies to @nalvietnam/avatar-cli@3.24.1.
See all versions of @nalvietnam/avatar-cli for other recorded verdicts.
- Review the evidence and your use of @nalvietnam/avatar-cli@3.24.1 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.