duplicate-checker

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill is purely instructional and contains no executable scripts or system commands. It functions as a set of rules for the AI model to follow when evaluating content uniqueness.\n- [NO_CODE]: No programming code, shell scripts, or software packages are included or required by this skill.\n- [PROMPT_INJECTION]: The skill processes untrusted 'new content' during analysis, which creates a surface for indirect prompt injection.\n
  • Ingestion points: Untrusted data enters the agent context via the 'new content' input described in SKILL.md.\n
  • Boundary markers: The instructions do not define delimiters or specific warnings to ignore embedded commands within the input content.\n
  • Capability inventory: The skill has no access to tools, subprocesses, file-writing, or network operations.\n
  • Sanitization: No escaping, validation, or filtering is performed on the untrusted input content.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 05:38 PM
Security Audit — agent-trust-hub — duplicate-checker