Security audit
Security checks for vulnerabilities and agentic risk
This is a markdown-only fact-checking skill whose behavior is disclosed and aligned with checking user-provided claims.
Installers should expect this skill to perform fact-checking and bias analysis, likely using web search for user-provided claims. It is mainly Chinese-language in presentation, so users who prefer another language may need to ask the agent to respond in that language.
Automatically triggering on any pasted text that contains statistics, claims, or assertions is ambiguous because many harmless inputs fit that pattern, including routine summaries, emails, or drafts. In an agent setting, this ambiguity can misroute user requests, invoke external-search behavior unexpectedly, and expose pasted content to processing paths the user did not clearly request.
Automatically triggering on any pasted text that contains statistics, claims, or assertions is ambiguous because many harmless inputs fit that pattern, including routine summaries, emails, or drafts. In an agent setting, this ambiguity can misroute user requests, invoke external-search behavior unexpectedly, and expose pasted content to processing paths the user did not clearly request.
Most of the skill content, examples, and output format are written in Chinese, which effectively biases the skill toward a specific language experience. Because the file does not explicitly offer the user a language or locale choice, this may violate a language/locale policy requiring opt-in rather than silent defaulting.
