@funny-booth/agent-core@0.1.46
AIエージェント起動CLI(共有ランチャー + MCPサーバー)
AI security review
scanned · by lpm-firewall-aiThe obfuscation and process/network signals warranted source review, but the inspected code showed no concrete attack behavior. The only additional report is tied to a failed user-invoked lookup and uses the configured portal client.
Decision evidence
public snapshot- package.json points the CLI and main entry to dist/index.js; its npm lifecycle hook is prepublishOnly for the build.