lead-import
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest lead data from external files or URLs, which presents a surface for indirect prompt injection if the data source contains malicious instructions intended to influence the agent's behavior during processing.
- Ingestion points: Lead data is accepted via a
source-fileorURLas defined in the skill's arguments. - Boundary markers: The instructions do not specify the use of clear delimiters or instructions to ignore embedded prompts within the lead data fields.
- Capability inventory: The skill can write data to CRM systems, trigger email nurture enrollments, send notifications via Slack/Email, and execute a local Python script (
approval-manager.py) to manage execution states. - Sanitization: While the skill performs rigorous format validation (email RFC 5322, phone E.164) and deduplication, it does not specifically sanitize the content for natural language instructions that might target the LLM.
Audit Metadata