guest-lecturer-program
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill documentation includes an installation command using npx to fetch and execute code from the author's official GitHub repository (https://github.com/peter-tu-zynkr/zynkr-skill-builder).
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from sources like Google Forms and Sheets to automate workflows, which introduces an injection surface.
- Ingestion points: Intake responses and lecturer tracking data are pulled from external Google Sheets (tracker_sheet_id) and Google Forms (intake_form_id).
- Boundary markers: While the skill instructs the agent to stop if configuration is missing, it lacks explicit delimiters or instruction-blocking markers for the data ingested from external sources.
- Capability inventory: The skill can update tracking sheets, generate contract drafts, compute financial payouts, and prepare settlement statements.
- Sanitization: The process relies on template-based artifacts and includes mandatory review checkpoints by legal and finance departments for final validation.
Audit Metadata