deep-interview

Pass

Audited by Gen Agent Trust Hub on Aug 11, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill facilitates structured communication and local documentation. It does not access sensitive credentials, perform network requests, or execute arbitrary shell commands.- [PROMPT_INJECTION]: Indirect prompt injection surface analysis:
  • Ingestion points: User-provided project descriptions via the skill argument (SKILL.md).
  • Boundary markers: The prompt templates do not use specific delimiters to isolate user input from instructions.
  • Capability inventory: Writing markdown specification files to the .omp/specs/ directory and suggesting the execution of other skills.
  • Sanitization: The skill does not perform explicit validation or sanitization of the input idea before including it in the generated specification.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 11, 2026, 02:22 AM
Security Audit — agent-trust-hub — deep-interview