training-client-team
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data through the 'Required Input' section—including client names, goals, and contact information—and substitutes these directly into a workbook template, creating a potential path for malicious instructions in the input to influence the agent's output.
- Ingestion points: User-provided strings for the 'Required Input' fields in SKILL.md.
- Boundary markers: Absent; user inputs are directly substituted into template brackets without delimiters.
- Capability inventory: Text generation and document formatting for a participant workbook. No network access, file writing, or subprocess execution capabilities were identified in the skill instructions.
- Sanitization: Absent; the skill does not include instructions for the agent to validate, escape, or filter the user-provided content before interpolation.
Audit Metadata