setup-outreach-campaign

Pass

Audited by Gen Agent Trust Hub on Mar 31, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXFILTRATION]: The skill transmits user-provided lead lists and campaign configuration to the official Smartlead API endpoint (https://server.smartlead.ai/api/v1). This data movement is the intended core functionality of the skill.
  • [CREDENTIALS_UNSAFE]: Secret management is handled via the SMARTLEAD_API_KEY environment variable. The skill correctly instructs the user to export the key rather than hardcoding it, which is consistent with secure practices.
  • [PROMPT_INJECTION]: The skill processes untrusted lead data from CSV or JSON sources which could host indirect prompt injection.
  • Ingestion points: Lead list data parsed in Step 4a.
  • Boundary markers: Absent in the prompt construction instructions.
  • Capability inventory: Network requests (POST) to Smartlead API for campaign creation and lead ingestion.
  • Sanitization: No automated sanitization is defined, but a human-in-the-loop (HITL) checkpoint is enforced in Step 5, requiring user approval of the email sequence before any data is sent to the sequence API.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 31, 2026, 09:58 PM