cold-email-outreach
Pass
Audited by Gen Agent Trust Hub on Mar 6, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references and interacts with several well-known technology services, including Smartlead (server.smartlead.ai), OpenRouter (openrouter.ai), and Cloudflare. These interactions are consistent with the skill's intended purpose and target established official domains.
- [COMMAND_EXECUTION]: Includes implementation examples using
curlto communicate with the Smartlead API. These examples correctly utilize environment variables for sensitive information like API keys, avoiding the use of hardcoded credentials. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted enrichment data through an AI node for email personalization.
- Ingestion points: Untrusted data enters the workflow via the
{{enrichment_summary}}variable in the n8n JSON snippet within SKILL.md. - Boundary markers: The prompt template does not use delimiters or explicit instructions to prevent the model from obeying instructions potentially embedded in the enrichment research.
- Capability inventory: The skill has the capability to generate email content and interact with the Smartlead API via
curlcommands. - Sanitization: There is no evidence of validation or filtering of the enrichment data before it is interpolated into the LLM prompt.
Audit Metadata