linkedin-abm-monthly-report

Pass

Audited by Gen Agent Trust Hub on Jul 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the weasyprint command-line utility to convert generated HTML reports into PDF documents for user delivery.
  • [EXTERNAL_DOWNLOADS]: The skill references branded templates and report assets located in the official ZENABM/linkedin-abm-skills repository on GitHub.
  • [PROMPT_INJECTION]: The skill processes untrusted data from external sources, presenting an indirect prompt injection surface.
  • Ingestion points: Data is retrieved from LinkedIn Ads via get_linkedin_metrics and get_creative_performance, as well as CRM deal information from HubSpot.
  • Boundary markers: The instructions do not define explicit delimiters or warnings to ignore instructions embedded within the ingested campaign names or company data.
  • Capability inventory: The skill has the capability to execute shell commands (weasyprint) and perform file system operations (writing HTML and PDF files).
  • Sanitization: No specific sanitization or escaping protocols are mentioned for data interpolated into the HTML report structure.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 28, 2026, 08:57 PM
Security Audit — agent-trust-hub — linkedin-abm-monthly-report