salesloft-core-workflow-a

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted prospect data which could potentially contain malicious instructions intended to influence the agent's behavior.
  • Ingestion points: Prospect names and emails are processed in the bulkEnroll and findOrCreatePerson functions in SKILL.md, typically sourced from external CSV files.
  • Boundary markers: The instructions do not define delimiters or provide specific prompts to the agent to ignore instructions embedded within the prospect data.
  • Capability inventory: The skill uses axios to perform authenticated network requests (GET and POST) to the SalesLoft REST API.
  • Sanitization: No input validation or sanitization is performed on the prospect data before it is incorporated into the API workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 03:02 PM
Security Audit — agent-trust-hub — salesloft-core-workflow-a