genpark-automated-email-marketing-agent

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill's installation instructions include cloning a repository from GitHub (github.com/alphaparkinc/genpark-automated-email-marketing-agent-skill.git). As this is the official repository for the skill authored by Aradotso (ara.so), it is considered a vendor-owned resource.
  • [PROMPT_INJECTION]: The skill implements a feature to process external data from CSV files (data/contacts.csv) which is then used by an LLM to generate personalized email content. This design presents an indirect prompt injection surface where malicious content within the contact data could potentially influence the agent's behavior or the generated output.
  • Ingestion points: AudienceSegmenter.load_from_csv in SKILL.md loads recipient profiles from a local file.
  • Boundary markers: No explicit delimiters or instructions to ignore instructions embedded in the data were found in the provided implementation examples.
  • Capability inventory: CampaignExecutor.launch_campaign provides the ability to perform network operations by sending emails via the SendGrid API.
  • Sanitization: The provided examples do not demonstrate sanitization or validation of the input fields (e.g., first_name, company) before they are used for content generation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 12:37 AM
Security Audit — agent-trust-hub — genpark-automated-email-marketing-agent