course-certification-pipeline

Pass

Audited by Gen Agent Trust Hub on Apr 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes system commands such as curl, pdfinfo, and pdftoppm to download, inspect, and convert certificate PDF files.
  • [EXTERNAL_DOWNLOADS]: Fetches certificate files from external domains including Skilljar verification endpoints.
  • [DATA_EXFILTRATION]: Automates the publication of certification details and media to LinkedIn as part of its primary intended workflow.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes content from external websites which presents a theoretical ingestion surface. Ingestion points: Skilljar course content and LinkedIn profile pages. Boundary markers: Absent. Capability inventory: Shell command execution and browser automation via CDP. Sanitization: No explicit content filtering described for ingested platform data.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 7, 2026, 12:56 PM