summarize-interview

Pass

Audited by Gen Agent Trust Hub on Mar 16, 2026

Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious instructions, obfuscation, or dangerous commands were detected in the skill instructions. The content is focused on legitimate UX research summarization tasks.\n- [NO_CODE]: The skill is entirely instruction-based and does not include any Python or Node.js code, scripts, or binary files, which minimizes its attack surface.\n- [PROMPT_INJECTION]: The skill processes untrusted user-provided interview transcripts, creating an indirect prompt injection surface. However, since the skill has no destructive capabilities, the security risk is negligible.\n
  • Ingestion points: User-provided transcripts (text or file) as described in SKILL.md.\n
  • Boundary markers: Absent; the skill does not include specific delimiters or warnings to ignore instructions embedded within the transcript text.\n
  • Capability inventory: None; the skill is limited to text generation and summarization without access to system tools, network, or file writing.\n
  • Sanitization: Absent; the skill does not perform validation or filtering on the transcript content before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 16, 2026, 12:08 AM