lov-fact-check

Pass

Audited by Gen Agent Trust Hub on Aug 24, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data (user claims and web search results), which constitutes a potential attack surface for indirect prompt injection.
  • Ingestion points: User-submitted text and external content retrieved via web/search tools as specified in the 'Workflow' section of SKILL.md.
  • Boundary markers: The instructions require the agent to restate and clarify claims before searching, which serves as a logical boundary. It also mandates the use of evidence labels (e.g., [一手], [二手]) to categorize findings.
  • Capability inventory: The skill uses web/search capabilities to retrieve data.
  • Sanitization: The instructions focus on critical evaluation of source quality and credibility rather than automated sanitization of external text.
  • [COMMAND_EXECUTION]: The README.md file documents a standard installation procedure for the platform environment.
  • Evidence: npx skills add fact-check -g -y is provided as a user-facing setup command.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 24, 2026, 12:55 AM
Security Audit — agent-trust-hub — lov-fact-check