interview-framework

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection through its dialogue-based requirement gathering process. 1. Ingestion points: Direct user input via triggers such as 'interview me' or 'clarify the request', and persistent state in .progress.md (SKILL.md). 2. Boundary markers: No explicit delimiters or 'ignore' instructions are provided to separate user-supplied content from the skill's operational logic. 3. Capability inventory: The skill has the ability to read and write to various local files including .progress.md, .planning/intake/*.md, and .v7/decisions/*.json (algorithm.md). 4. Sanitization: There is no defined process for validating, escaping, or filtering user input before it is summarized or written to disk.
  • [NO_CODE]: The skill consists entirely of instructional markdown and reference documentation. No executable scripts, binary files, or remote dependencies are included in the provided files.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 08:44 AM
Security Audit — agent-trust-hub — interview-framework