@vpxa/aikit@0.1.384
Local-first AI developer toolkit — knowledge base, code analysis, context management, and developer tools for LLM agents
AI Security Review
scanned 3mo ago · by lpm-firewall-aiThe lifecycle hook mutates AI-agent control surfaces, but only in a narrowly guarded first-party managed-update flow and only when migrating an existing AI Kit registration. Source inspection found no credential theft, exfiltration, stealth broad config takeover, or foreign endpoint.
Decision evidence
public snapshot- package.json runs postinstall
- bin/postinstall-managed-launcher.mjs conditionally imports CLI
- packages/cli/dist/index.js migrates detected IDE MCP entries and modifies Claude settings/hooks
- packages/server/dist/startup-maintenance-D7tKqECM.js checks npm and installs newer versions
- Postinstall is a no-op outside ~/.aikit/versions/v*-staging/package
- Migration only replaces an existing 'AI Kit' MCP entry
- No credential collection or external data-exfiltration path was confirmed
- Registry request is package-aligned update behavior
Source & flagged code
13 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgPackage source references child process execution.
packages/tools/dist/index.jsView on unpkg · L1Findings
Affected versions and remediation
This report applies to @vpxa/aikit@0.1.384.
See all versions of @vpxa/aikit for other recorded verdicts.
- Review the evidence and your use of @vpxa/aikit@0.1.384 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.