list-architect
Fail
Audited by Gen Agent Trust Hub on Jul 28, 2026
Risk Level: CRITICAL
Full Analysis
- [PROMPT_INJECTION]: No evidence of prompt injection or safety bypass attempts was found. All provided prompts are task-oriented and used for data classification or summarization within the defined list-building workflow.- [EXTERNAL_DOWNLOADS]: The skill references numerous legitimate B2B data sources and directories. While saasmag.com was flagged by automated scanning, it is a known industry publication relevant to the skill's list-building focus and does not involve automatic execution of remote code.- [DATA_EXFILTRATION]: No patterns indicative of credential harvesting or data exfiltration were detected. The skill provides instructional guidance for using established third-party platforms such as Clay, Sales Navigator, and HubSpot.- [INDIRECT_PROMPT_INJECTION]: The skill documentation describes workflows that ingest external data, such as company summaries, for processing by AI models. While this represents a theoretical attack surface, the instructions include specific output constraints and structured prompts that serve as basic mitigations.
Recommendations
- Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata