القائمة
Update interpersonal familiarity, trust, liking, obligation, conflict, and shared history.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
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 | relationships |
| description | Update interpersonal familiarity, trust, liking, obligation, conflict, and shared history. |
| script | scripts/update_relationships.py |
Maintain social continuity between agents. This skill turns repeated interaction into relationship state instead of treating every encounter as new.
Research basis: references/research_basis.md.
Read recent social interaction and prior relationship state, update bounded familiarity, trust, liking, obligation, conflict, respect, and optional opinion-influence weights, then write state/relationships.json and notable social events.
Use this skill after conversations, cooperation, conflict, promises, favors, gifts, avoidance, betrayal, or repeated co-presence.
state/observation.txt, state/observation_ctx.json, state/relationships.json, state/memory.jsonl, state/emotion.json, and profile context if present.state/social_events.jsonl.If deterministic baseline is preferred, run scripts/update_relationships.py first, then optionally refine subtle social interpretation with LLM reasoning.
Write state/relationships.json. Optionally append state/social_events.jsonl.
{
"people": {
"alice": {
"familiarity": 0.62,
"trust": 0.71,
"liking": 0.55,
"obligation": 0.2,
"conflict": 0.05,
"respect": 0.58,
"influence_weight": 0.34,
"last_interaction": "brief friendly chat at cafe",
"shared_history_tags": ["cafe", "work"]
}
}
}
Trust and conflict should change more slowly than momentary emotion. A single event can strongly affect a relationship only if it is high-stakes, public, repeated, or identity-relevant.
Consensus note: if multiple trusted people provide opinions about the same uncertain issue, later cognition or media_literacy can average those opinions using relationship influence weights. This follows DeGroot-style consensus as a simple social influence baseline, not as a claim that real groups always converge.