Security audit
Security checks for vulnerabilities and agentic risk
This is a markdown-only skill for summarizing a public RSS feed with financial-education commentary, with some persona-style output that users should understand is simulated.
Install only if you want this specific Trump-post/RSS finance-education workflow. Treat the Trump-style responses as simulated commentary, not authentic statements or investment advice, and verify any market-relevant claims against primary sources before acting.
The manifest description says the skill summarizes Trump-related public posts and chats in Trump's style, but it does not specify concrete trigger phrases, scope limits, or exclusion conditions. In a manifest file, this kind of broad description can overlap with many general requests about Trump, finance, or conversation style and may cause unintended invocation.
The skill description says it will converse in Trump's voice without requiring user opt-in, which can cause unwanted impersonation or deceptive framing. In a politically sensitive and finance-related context, forced persona use may mislead users about source, endorsement, or authenticity and can undermine informed consent.
This section mandates 'Trump dialogue mode' whenever users ask for Trump's views without asking for original evidence, removing user choice and enforcing impersonation by default. Because the skill discusses markets and investing, that persona can amplify persuasion and create confusion between educational analysis and simulated authoritative statements.
The output template requires a Trump-style opening, which operationalizes impersonation at the formatting layer and makes it difficult for downstream systems or users to avoid. In a finance-adjacent skill, mandatory persona framing can increase social engineering risk, political manipulation concerns, and user misunderstanding about provenance.
No suspicious patterns detected.
