public-opinion-executor-skill

Generate closed-book Chinese public-opinion risk analysis from event-packs with evidence-tier reasoning.

3|1|Updated Apr 10, 2026
One-click install
npx skills add https://github.com/zdy85730/learn-skill --skill public-opinion-executor-skill
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: public-opinion-executor-skill
Source: https://github.com/zdy85730/learn-skill/tree/main/scenarios/public-opinion
Command: npx skills add https://github.com/zdy85730/learn-skill --skill public-opinion-executor-skill

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you assess Chinese public-opinion risk from an event-pack while preventing hallucinations by enforcing strict fact/signal separation and closed-book constraints.

Core Features & Use Cases

  • Closed-book execution: Produces risk analysis using only the provided case pack and bundled rules (no external corpus, memory, or judge gold).
  • Evidence-tier disciplined reasoning: Treats A1/A2 as verified anchors, B as context, C as signal-only, and applies abstention/uncertainty rules when evidence is weak or conflicting.
  • Public-opinion risk outputs: Generates structured outputs including risk_level, core_claims, signals, evidence-by-tier, uncertainty, recommended_actions, abstain, and confidence—suitable for downstream judging.

Quick Start

Use the public-opinion-executor-skill to analyze the attached current case pack by producing the required JSON fields plus a short summary that clearly distinguishes verified official boundaries, confirmed facts, and observed social signals.

Frequently Asked Questions about public-opinion-executor-skill

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I analyze public-opinion risk without hallucinating on weak signals?

You can analyze public-opinion risk without hallucinating by using closed-book execution that enforces strict fact/signal separation, treating verified facts as anchors and applying abstention rules when only weak signals exist.

What is evidence-tier aware reasoning for rumor clarification?

Evidence-tier aware reasoning for rumor clarification separates verified facts from community signals by assigning evidence tiers, treating A1/A2 as anchors, B as context, and C as signal-only to prevent unverified claims from driving conclusions.

How do I generate a structured public-opinion risk report from an event-pack?

Generate a structured public-opinion risk report by feeding the event-pack to a closed-book executor, which outputs risk level, core claims, evidence by tier, uncertainty, recommended actions, and confidence scores.

When should I abstain from making a public-opinion risk judgment?

You should abstain from making a public-opinion risk judgment when only C or D tier signals exist, as evidence-tier disciplined reasoning requires explicit uncertainty and abstention when evidence is weak or conflicting.

Does closed-book public-opinion analysis require external corpus or memory access?

No, closed-book public-opinion analysis does not require external corpus, memory, or hidden judge materials, as it strictly constrains reasoning to only the provided event-pack and bundled rules to prevent data leakage.

Can I detect coordination-style rumor spread using event-pack analysis?

Yes, you can detect coordination-style rumor spread using event-pack analysis by isolating verified official boundaries from observed social signals and sarcasm sentiment to map coordinatedspread patterns at a general-practitioner level.