discovery-interview-prep

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious patterns or security vulnerabilities were found. The skill operates as a structured prompt template.
  • [PROMPT_INJECTION]: The skill processes external text provided by the user in Step 0 (e.g., Reddit reviews). While this represents a data ingestion surface for indirect prompt injection, the risk is negligible as the skill lacks any high-risk capabilities such as network access or command execution.
  • Ingestion points: Step 0 context gathering (SKILL.md).
  • Boundary markers: None explicitly defined.
  • Capability inventory: No subprocess calls, network operations, or file system access detected.
  • Sanitization: No explicit filtering of user-provided context.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 12:54 AM