linkedin-hook-generator
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is instructed to read content from local configuration files (.liftli/voice.md and ~/.liftli/voice.md). While these serve as ingestion points for external data without explicit boundary markers or sanitization, the skill has no dangerous tools or capabilities (such as network exfiltration or file-writing tools) that could be exploited by a malicious payload in those files.
- [EXTERNAL_DOWNLOADS]: The skill references the vendor's official MCP server (mcp.liftli.ai) and provides user instructions for installation of vendor-owned skills. These are legitimate resources belonging to the skill author (liftli-ai) and represent normal application functionality.
Audit Metadata