cold-outreach-weapon
Pass
Audited by Gen Agent Trust Hub on Jul 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists entirely of Markdown-based playbooks, guides, and templates designed to educate users on B2B sales workflows. No executable scripts, shell commands, or malicious behavioral patterns were found across the 38 analyzed files.
- [EXTERNAL_DOWNLOADS]: The skill references numerous external URLs for research grounding and tool documentation (e.g., official Google Workspace documentation, Apollo, Smartlead, and various industry blogs). These are informational references only and do not involve automated software installation or remote code execution.
- [DATA_EXFILTRATION]: The skill provides instructions on how to handle prospect lead data (such as names and LinkedIn URLs) for sales outreach. It does not contain any code or instructions that attempt to access or exfiltrate sensitive local user data, environment variables, or private credentials.
- [PROMPT_INJECTION]: The instructions include domain-specific 'non-negotiable' principles and workflows (e.g., ICP tightness, deliverability setup). These are legitimate instructional guardrails and do not attempt to bypass safety filters, override system prompts, or induce unrestricted AI behavior.
Audit Metadata