菜单
Track information exposure, source credibility, motivated reasoning, misinformation risk, and belief-update resistance.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Assess what actions are realistically available under environment, time, distance, access, money, body, and social constraints.
Maintain sleep-wake rhythm, circadian alertness, appetite rhythm, and chronotype-sensitive daily timing.
Produce emotion.json and intention.json from workspace context.
Model conversation intent, speech style, turn-taking, listening, repair, and nonverbal cues. Use before or after social interaction, dialogue, negotiation, apology, request, gossip, or conflict.
Apply cultural values, etiquette, rituals, symbols, taboos, and local meaning to perception and decisions.
Track sickness, pain, chronic condition, recovery, exercise, stress load, and long-term wellbeing.
| name | media_literacy |
| description | Track information exposure, source credibility, motivated reasoning, misinformation risk, and belief-update resistance. |
| script | scripts/update_media_literacy.py |
Model how agents process claims from media, friends, institutions, ads, rumors, and social platforms. This skill prevents agents from instantly accepting every statement and makes belief change depend on source credibility, prior beliefs, identity alignment, repetition, and warning/inoculation.
Research basis: references/research_basis.md.
Read observation, memory, relationships, identity, prior beliefs, and prior media state, then estimate claim credibility, misinformation risk, confirmation bias, inoculation strength, and write state/media_literacy.json plus optional belief update hints.
Use after news exposure, social media posts, rumors, advertising, political claims, health claims, scams, institutional announcements, warnings, debunks, or repeated claims.
state/observation.txt, state/memory.jsonl, state/relationships.json, state/identity.json, state/beliefs.json, and state/media_literacy.json if present.state/media_literacy.json.state/belief_update_hints.json for reflection/cognition to use.If deterministic baseline is preferred:
python skills/media_literacy/scripts/update_media_literacy.py --state-dir state --tick 120
Belief uptake is not just evidence:
acceptance_tendency =
source_credibility
+ evidence_quality
+ familiarity
+ identity_alignment
+ emotional_arousal
- misinformation_risk
- inoculation_strength
The output should not directly rewrite stable beliefs unless evidence is strong. It should provide a hint for reflection.
Always write state/media_literacy.json.
Optionally write state/belief_update_hints.json.
{
"_meta": {
"skill": "media_literacy",
"purpose": "Current information exposure assessment and belief-update caution."
},
"_summary": "A health claim from an unknown source has high misinformation risk.",
"current_claim": "A viral post says the medicine is dangerous.",
"source_type": "social_media",
"source_credibility": 0.32,
"evidence_quality": 0.18,
"familiarity": 0.61,
"identity_alignment": 0.44,
"emotional_arousal": 0.7,
"misinformation_risk": 0.78,
"inoculation_strength": 0.35,
"acceptance_tendency": 0.22,
"recommended_stance": "withhold belief and seek corroboration"
}